Inverse Conformational Selection in Lipid-Protein Binding.

Interest in lipid interactions with proteins and other biomolecules is emerging not only in fundamental biochemistry but also in the field of nanobiotechnology where lipids are commonly used, for example, in carriers of mRNA vaccines. The outward-facing components of cellular membranes and lipid nanoparticles, the lipid headgroups, regulate membrane interactions with approaching substances, such as proteins, drugs, RNA, or viruses. Because lipid headgroup conformational ensembles have not been experimentally determined in physiologically relevant conditions, an essential question about their interactions with other biomolecules remains unanswered: Do headgroups exchange between a few rigid structures, or fluctuate freely across a practically continuous spectrum of conformations? Here, we combine solid-state NMR experiments and molecular dynamics simulations from the NMRlipids Project to resolve the conformational ensembles of headgroups of four key lipid types in various biologically relevant conditions. We find that lipid headgroups sample a wide range of overlapping conformations in both neutral and charged cellular membranes, and that differences in the headgroup chemistry manifest only in probability distributions of conformations. Furthermore, the analysis of 894 protein-bound lipid structures from the Protein Data Bank suggests that lipids can bind to proteins in a wide range of conformations, which are not limited by the headgroup chemistry. We propose that lipids can select a suitable headgroup conformation from the wide range available to them to fit the various binding sites in proteins. The proposed inverse conformational selection model will extend also to lipid binding to targets other than proteins, such as drugs, RNA, and viruses.

Atomistic and Coarse Grain Simulations of the Cell Envelope of Gram-Negative Bacteria: What Have We Learned?

Bacterial membranes, and those of Gram-negative bacteria in particular, are some of the most biochemically diverse membranes known. They incorporate a wide range of lipid types and proteins of varying sizes, architectures, and functions. While simpler biological membranes have been the focus of myriad simulation studies over the years that have yielded invaluable details to complement, and often to direct, ongoing experimental studies, simulations of complex bacterial membranes have been slower to emerge. However, the past few years have seen tremendous activity in this area, leading to advances such as the development of atomistic and coarse-grain models of the lipopolysaccharide (LPS) component of the outer membrane that are compatible with widely used simulation codes. In this Account, we review our contributions to the field of molecular simulations of the bacterial cell envelope, including the development of models of both membranes and the cell wall of Gram-negative bacteria, with a predominant focus on E. coli. At the atomistic level, simulations of chemically accurate models of both membranes have revealed the tightly cross-linked nature of the LPS headgroups and have shown that penetration of solutes through these regions is not as straightforward as the route through phospholipids. The energetic differences between the two routes have been calculated. Simulations of native outer membrane proteins in LPS-containing membranes have shown that the conformational dynamics of the proteins is not only slower in LPS but also different compared to in simpler models of phospholipid bilayers. These chemically more complex and consequently biologically more relevant models are leading to details of conformational dynamics that were previously inaccessible from simulations. Coarse-grain models have enabled simulations of multiprotein systems on time scales of microseconds, leading to insights not only into the rates of protein and lipid diffusion but also into the trends in their respective directions of flow. We find that the motions of LPS molecules are highly correlated with each other but also with outer membrane proteins embedded within the membrane. We have shown that the two leaflets of the outer membrane exhibit communication, whereby regions of low disorder in one leaflet correspond to regions of high disorder in the other. The cell wall remains a comparatively neglected component, although models of the E. coli peptidoglycan are now emerging, particularly at the atomistic level. Our simulations of Braun's lipoprotein have shown that bending and tilting of this protein afford a degree of variability in the gap between the cell wall and the OM. The noncovalent interactions with the cell wall of proteins such as OmpA can further influence the width of this gap by extension or contraction of their linker domains. Overall we have shown that the dynamics of proteins, lipids, and other molecular species within the outer membrane cannot be approximated using simpler phospholipid bilayers, if one is addressing questions regarding the in vivo behavior of Gram-negative bacteria. These membranes have their own unique chemical characteristics that cannot be decoupled from their biological functions.

Braun's Lipoprotein Facilitates OmpA Interaction with the Escherichia coli Cell Wall.

Gram-negative bacteria such as Escherichia coli are protected by a complex cell envelope. The development of novel therapeutics against these bacteria necessitates a molecular level understanding of the structure-dynamics-function relationships of the various components of the cell envelope. We use atomistic MD simulations to reveal the details of covalent and noncovalent protein interactions that link the outer membrane to the aqueous periplasmic region. We show that the Braun's lipoprotein tilts and bends, and thereby lifts the cell wall closer to the outer membrane. Both monomers and dimers of the outer membrane porin OmpA can interact with peptidoglycan in the presence of Braun's lipoprotein, but in the absence of the latter, only dimers of OmpA show a propensity to form contacts with peptidoglycan. Our study provides a glimpse of how the molecular components of the bacterial cell envelope interact with each other to mediate cell wall attachment in E. coli.

Full-Length OmpA: Structure, Function, and Membrane Interactions Predicted by Molecular Dynamics Simulations.

OmpA is a multidomain protein found in the outer membranes of most Gram-negative bacteria. Despite a wealth of reported structural and biophysical studies, the structure-function relationships of this protein remain unclear. For example, it is still debated whether it functions as a pore, and the precise molecular role it plays in attachment to the peptidoglycan of the periplasm is unknown. The absence of a consensus view is partly due to the lack of a complete structure of the full-length protein. To address this issue, we performed molecular-dynamics simulations of the full-length model of the OmpA dimer proposed by Robinson and co-workers. The N-terminal domains were embedded in an asymmetric model of the outer membrane, with lipopolysaccharide molecules in the outer leaflet and phospholipids in the inner leaflet. Our results reveal a large dimerization interface within the membrane environment, ensuring that the dimer is stable over the course of the simulations. The linker is flexible, expanding and contracting to pull the globular C-terminal domain up toward the membrane or push it down toward the periplasm, suggesting a possible mechanism for providing mechanical stability to the cell. The external loops were more stabilized than was observed in previous studies due to the extensive dimerization interface and presence of lipopolysaccharide molecules in our outer-membrane model, which may have functional consequences in terms of OmpA adhesion to host cells. In addition, the pore-gating behavior of the protein was modulated compared with previous observations, suggesting a possible role for dimerization in channel regulation.

Interaction of the antimicrobial peptide polymyxin B1 with both membranes of E. coli: a molecular dynamics study.

Antimicrobial peptides are small, cationic proteins that can induce lysis of bacterial cells through interaction with their membranes. Different mechanisms for cell lysis have been proposed, but these models tend to neglect the role of the chemical composition of the membrane, which differs between bacterial species and can be heterogeneous even within a single cell. Moreover, the cell envelope of Gram-negative bacteria such as E. coli contains two membranes with differing compositions. To this end, we report the first molecular dynamics simulation study of the interaction of the antimicrobial peptide, polymyxin B1 with complex models of both the inner and outer membranes of E. coli. The results of >16 microseconds of simulation predict that polymyxin B1 is likely to interact with the membranes via distinct mechanisms. The lipopeptides aggregate in the lipopolysaccharide headgroup region of the outer membrane with limited tendency for insertion within the lipid A tails. In contrast, the lipopeptides readily insert into the inner membrane core, and the concomitant increased hydration may be responsible for bilayer destabilization and antimicrobial function. Given the urgent need to develop novel, potent antibiotics, the results presented here reveal key mechanistic details that may be exploited for future rational drug development.

The NorM MATE transporter from N. gonorrhoeae: insights into drug and ion binding from atomistic molecular dynamics simulations.

The multidrug and toxic compound extrusion transporters extrude a wide variety of substrates out of both mammalian and bacterial cells via the electrochemical gradient of protons and cations across the membrane. The substrates transported by these proteins include toxic metabolites and antimicrobial drugs. These proteins contribute to multidrug resistance in both mammalian and bacterial cells and are therefore extremely important from a biomedical perspective. Although specific residues of the protein are known to be responsible for the extrusion of solutes, mechanistic details and indeed structures of all the conformational states remain elusive. Here, we report the first, to our knowledge, simulation study of the recently resolved x-ray structure of the multidrug and toxic compound extrusion transporter, NorM from Neisseria gonorrhoeae (NorM_NG). Multiple, atomistic simulations of the unbound and bound forms of NorM in a phospholipid lipid bilayer allow us to identify the nature of the drug-protein/ion-protein interactions, and secondly determine how these interactions contribute to the conformational rearrangements of the protein. In particular, we identify the molecular rearrangements that occur to enable the Na(+) ion to enter the cation-binding cavity even in the presence of a bound drug molecule. These include side chain flipping of a key residue, GLU-261 from pointing toward the central cavity to pointing toward the cation binding side when bound to a Na(+) ion. Our simulations also provide support for cation binding in the drug-bound and apo states of NorM_NG.

The structural basis for endotoxin-induced allosteric regulation of the Toll-like receptor 4 (TLR4) innate immune receptor.

As part of the innate immune system, Toll-like receptor 4 (TLR4) recognizes bacterial cell surface lipopolysaccharide (LPS) by forming a complex with a lipid-binding co-receptor, MD-2. In the presence of agonist, TLR4·MD-2 dimerizes to form an active receptor complex, leading to initiation of intracellular inflammatory signals. TLR4 is of great biomedical interest, but its pharmacological manipulation is complicated because even subtle variations in the structure of LPS can profoundly impact the resultant immunological response. Here, we use atomically detailed molecular simulations to gain insights into the nature of the molecular signaling mechanism. We first demonstrate that MD-2 is extraordinarily flexible. The "clamshell-like" motions of its ß-cup fold enable it to sensitively match the volume of its hydrophobic cavity to the size and shape of the bound lipid moiety. We show that MD-2 allosterically transmits this conformational plasticity, in a ligand-dependent manner, to a phenylalanine residue (Phe-126) at the cavity mouth previously implicated in TLR4 activation. Remarkably, within the receptor complex, we observe spontaneous transitions between active and inactive signaling states of Phe-126, and we confirm that Phe-126 is indeed the "molecular switch" in endotoxic signaling.

Conformational dynamics and membrane interactions of the E. coli outer membrane protein FecA: a molecular dynamics simulation study.

The TonB-dependent transporters mediate high-affinity binding and active transport of a variety of substrates across the outer membrane of Escherichia coli. The substrates transported by these proteins are large, scarce nutrients that are unable to gain entry into the cell by passive diffusion across the complex, asymmetric bilayer that constitutes the outer membrane. Experimental studies have identified loop regions that are essential for the correct functioning of these proteins. A number of these loops have been implicated in ligand binding. We report the first simulations of an E. coli outer membrane protein in an asymmetric model membrane that incorporates lipopolysaccharide (LPS) molecules. Comparative simulations of the apo and holo forms of the TonB-dependent transporter FecA in different membrane models enable us to identify the nature of the LPS-protein interactions and determine how these interactions impact upon the conformational dynamics of this protein. In particular, our simulations provide molecular-level insights into the influence of the environment and ligand on the dynamics of the functionally important loops of FecA. In addition, we provide insights into the nature of the protein-ligand interactions and ligand induced conformational change in FecA.

Stability and membrane interactions of an autotransport protein: MD simulations of the Hia translocator domain in a complex membrane environment.

Hia is a trimeric autotransporter found in the outer membrane of Haemphilus influenzae. The X-ray structure of Hia translocator domain revealed each monomer to consist of an α-helix connected via a loop to a 4-stranded ß-sheet, thus the topology of the trimeric translocator domain is a 12-stranded ß-barrel containing 3 α-helices that protrude from the mouth of the ß-barrel into the extracellular medium. Molecular dynamics simulations of the Hia monomer and trimer have been employed to explore the interactions between the helices, ß-barrel and connecting loops that may contribute to the stability of the trimer. In simulations of the Hia monomer we show that the central α-helix may stabilise the fold of the 4-stranded ß-sheet. In simulations of the Hia trimer, a H-bond network involving residues in the ß-barrel, α-helices and loops has been identified as providing stability for the trimeric arrangement of the monomers. Glutamine residues located in the loops connecting the α-helices to the ß-barrel are orientated in a triangular arrangement such that each forms 2 hydrogen bonds to each of the corresponding glutamines in the other loops. In the absence of the loops, the ß-barrel becomes distorted. Simulations show that while the trimeric translocator domain ß-barrel is inherently flexible, it is unlikely to accommodate the passenger domain in a folded conformation. Simulations of Hia in an asymmetric model of the outer membrane have revealed membrane-protein interactions that anchor the protein within its native membrane environment.

Overlay databank unlocks data-driven analyses of biomolecules for all.

Tools based on artificial intelligence (AI) are currently revolutionising many fields, yet their applications are often limited by the lack of suitable training data in programmatically accessible format. Here we propose an effective solution to make data scattered in various locations and formats accessible for data-driven and machine learning applications using the overlay databank format. To demonstrate the practical relevance of such approach, we present the NMRlipids Databank-a community-driven, open-for-all database featuring programmatic access to quality-evaluated atom-resolution molecular dynamics simulations of cellular membranes. Cellular membrane lipid composition is implicated in diseases and controls major biological functions, but membranes are difficult to study experimentally due to their intrinsic disorder and complex phase behaviour. While MD simulations have been useful in understanding membrane systems, they require significant computational resources and often suffer from inaccuracies in model parameters. Here, we demonstrate how programmable interface for flexible implementation of data-driven and machine learning applications, and rapid access to simulation data through a graphical user interface, unlock possibilities beyond current MD simulation and experimental studies to understand cellular membranes. The proposed overlay databank concept can be further applied to other biomolecules, as well as in other fields where similar barriers hinder the AI revolution.

Single-stranded DNA within nanopores: conformational dynamics and implications for sequencing; a molecular dynamics simulation study.

Engineered protein nanopores, such as those based on α-hemolysin from Staphylococcus aureus have shown great promise as components of next-generation DNA sequencing devices. However, before such protein nanopores can be used to their full potential, the conformational dynamics and translocation pathway of the DNA within them must be characterized at the individual molecule level. Here, we employ atomistic molecular dynamics simulations of single-stranded DNA movement through a model α-hemolysin pore under an applied electric field. The simulations enable characterization of the conformations adopted by single-stranded DNA, and allow exploration of how the conformations may impact on translocation within the wild-type model pore and a number of mutants. Our results show that specific interactions between the protein nanopore and the DNA can have a significant impact on the DNA conformation often leading to localized coiling, which in turn, can alter the order in which the DNA bases exit the nanopore. Thus, our simulations show that strategies to control the conformation of DNA within a protein nanopore would be a distinct advantage for the purposes of DNA sequencing.

Toward a detailed description of the pathways of allosteric communication in the GroEL chaperonin through atomistic simulation.

GroEL, along with its coprotein GroES, is essential for ensuring the correct folding of unfolded or newly synthesized proteins in bacteria. GroEL is a complex, allosteric molecule, composed of two heptameric rings stacked back to back, that undergoes large structural changes during its reaction cycle. These structural changes are driven by the cooperative binding and subsequent hydrolysis of ATP, by GroEL. Despite numerous previous studies, the precise mechanisms of allosteric communication and the associated structural changes remain elusive. In this paper, we describe a series of all-atom, unbiased, molecular dynamics simulations over relatively long (50-100 ns) time scales of a single, isolated GroEL subunit and also a heptameric GroEL ring, in the presence and absence of ATP. Combined with results from a distance restraint-biased simulation of the single ring, the atomistic details of the earliest stages of ATP-driven structural changes within this complex molecule are illuminated. Our results are in broad agreement with previous modeling studies of isolated subunits and with a coarse-grained, forcing simulation of the single ring. These are the first reported all-atom simulations of the GroEL single-ring complex and provide a unique insight into the role of charged residues K80, K277, R284, R285, and E388 at the subunit interface in transmission of the allosteric signal. These simulations also demonstrate the feasibility of performing all-atom simulations of very large systems on sufficiently long time scales on typical high performance computing facilities to show the origins of the earliest events in biologically relevant processes.

Molecular Dynamics Simulation of the Interaction of Two Linear Battacin Analogs with Model Gram-Positive and Gram-Negative Bacterial Cell Membranes.

Antimicrobial peptides (AMPs) are a potential solution to the increasing threat of antibiotic resistance, but successful design of active but nontoxic AMPs requires understanding their mechanism of action. Molecular dynamics (MD) simulations can provide atomic-level information regarding how AMPs interact with the cell membrane. Here, we have used MD simulations to study two linear analogs of battacin, a naturally occurring cyclic, lipidated, nonribosomal AMP. Like battacin, these analogs are active against Gram-negative multidrug resistant and Gram-positive bacteria, but they are less toxic than battacin. Our simulations show that this activity depends upon a combination of positively charged and hydrophobic moieties. Favorable interactions with negatively charged membrane lipid head groups drive association with the membrane and insertion of hydrophobic residues, and the N-terminal lipid anchors the peptides to the membrane surface. Both effects are required for stable membrane binding.

Molecular determinants of binding of non-oxime bispyridinium nerve agent antidote compounds to the adult muscle nAChR.

Organophosphorus nerve agents (NAs) are the most lethal chemical warfare agents and have been used by state and non-state actors since their discovery in the 1930s. They covalently modify acetylcholinesterase, preventing the breakdown of acetylcholine (ACh) with subsequent loss of synaptic transmission, which can result in death. Despite the availability of several antidotes for OPNA exposure, none directly targets the nicotinic acetylcholine receptor (nAChR) mediated component of toxicity. Non-oxime bispyridinium compounds (BPDs) have been shown previously to partially counteract the effects of NAs at skeletal muscle tissue, and this has been attributed to inhibition of the muscle nAChR. Functional data indicate that, by increasing the length of the alkyl linker between the pyridinium moieties of BPDs, the antagonistic activity at nAChRs can be improved. Molecular dynamics simulations of the adult muscle nAChR in the presence of BPDs identified key residues likely to be involved in binding. Subsequent two-electrode voltage clamp recordings showed that one of the residues, εY131, acts as an allosteric determinant of BPD binding, and that longer BPDs have a greater stabilizing effect on the orthosteric loop C than shorter ones. The work reported will inform future design work on novel antidotes for treating NA exposure.

Stability and membrane orientation of the fukutin transmembrane domain: a combined multiscale molecular dynamics and circular dichroism study.

The N-terminal domain of fukutin-I has been implicated in the localization of the protein in the endoplasmic reticulum and Golgi Apparatus. It has been proposed to mediate this through its interaction with the thinner lipid bilayers found in these compartments. Here we have employed multiscale molecular dynamics simulations and circular dichroism spectroscopy to explore the structure, stability, and orientation of the short 36-residue N-terminus of fukutin-I (FK1TMD) in lipids with differing tail lengths. Our results show that FK1TMD adopts a stable helical conformation in phosphatidylcholine lipids when oriented with its principal axis perpendicular to the bilayer plane. The stability of the helix is largely insensitive to the lipid tail length, preventing hydrophobic mismatch by virtue of its mobility and ability to tilt within the lipid bilayers. This suggests that changes in FK1TMD tilt in response to bilayer properties may be implicated in the regulation of its trafficking. Coarse-grained simulations of the complex Golgi membrane suggest the N-terminal domain may induce the formation of microdomains in the surrounding membrane through its preferential interaction with 1,2-dipalmitoyl-sn-glycero-3-phosphatidylinositol 4,5-bisphosphate lipids.

Molecular Dynamics Simulation of Proteins.

Molecular dynamics simulations allow the conformational motion of a molecule such as a protein to be followed over time at atomic-level detail. Several choices need to be made prior to running a simulation, including the software, which molecules to include in the simulation, and the force field used to describe their behavior. Guidance on making these choices and other important aspects of running MD simulations is outlined here.

Headgroup Structure and Cation Binding in Phosphatidylserine Lipid Bilayers.

Phosphatidylserine (PS) is a negatively charged lipid type commonly found in eukaryotic membranes, where it interacts with proteins via nonspecific electrostatic interactions as well as via specific binding. Moreover, in the presence of calcium ions, PS lipids can induce membrane fusion and phase separation. Molecular details of these phenomena remain poorly understood, partly because accurate models to interpret the experimental data have not been available. Here we gather a set of previously published experimental NMR data of C-H bond order parameter magnitudes, |SCH|, for pure PS and mixed PS:PC (phosphatidylcholine) lipid bilayers and augment this data set by measuring the signs of SCH in the PS headgroup using S-DROSS solid-state NMR spectroscopy. The augmented data set is then used to assess the accuracy of the PS headgroup structures in, and the cation binding to, PS-containing membranes in the most commonly used classical molecular dynamics (MD) force fields including CHARMM36, Lipid17, MacRog, Slipids, GROMOS-CKP, Berger, and variants. We show large discrepancies between different force fields and that none of them reproduces the NMR data within experimental accuracy. However, the best MD models can detect the most essential differences between PC and PS headgroup structures. The cation binding affinity is not captured correctly by any of the PS force fields-an observation that is in line with our previous results for PC lipids. Moreover, the simulated response of the PS headgroup to bound ions can differ from experiments even qualitatively. The collected experimental data set and simulation results will pave the way for development of lipid force fields that correctly describe the biologically relevant negatively charged membranes and their interactions with ions. This work is part of the NMRlipids open collaboration project ( nmrlipids.blogspot.fi ).

Evidence for phospholipid export from the bacterial inner membrane by the Mla ABC transport system.

The Mla pathway is believed to be involved in maintaining the asymmetrical Gram-negative outer membrane via retrograde phospholipid transport. The pathway is composed of three components: the outer membrane MlaA-OmpC/F complex, a soluble periplasmic protein, MlaC, and the inner membrane ATPase, MlaFEDB complex. Here, we solve the crystal structure of MlaC in its phospholipid-free closed apo conformation, revealing a pivoting ß-sheet mechanism that functions to open and close the phospholipid-binding pocket. Using the apo form of MlaC, we provide evidence that the inner-membrane MlaFEDB machinery exports phospholipids to MlaC in the periplasm. Furthermore, we confirm that the phospholipid export process occurs through the MlaD component of the MlaFEDB complex and that this process is independent of ATP. Our data provide evidence of an apparatus for lipid export away from the inner membrane and suggest that the Mla pathway may have a role in anterograde phospholipid transport.

A Thermodynamic Funnel Drives Bacterial Lipopolysaccharide Transfer in the TLR4 Pathway.

The Gram-negative bacterial outer membrane contains lipopolysaccharide, which potently stimulates the mammalian innate immune response. This involves a relay of specialized complexes culminating in transfer of lipopolysaccharide from CD14 to Toll-like receptor 4 (TLR4) and its co-receptor MD-2 on the cell surface, leading to activation of downstream inflammatory responses. In this study we develop computational models to trace the TLR4 cascade in near-atomic detail. We demonstrate through rigorous thermodynamic calculations that lipopolysaccharide molecules traversing the receptor cascade fall into a thermodynamic funnel. An affinity gradient for lipopolysaccharide is revealed upon extraction from aggregates or realistic bacterial outer membrane models and transfer through CD14 to the terminal TLR4/MD-2 receptor-co-receptor complex. We subsequently assemble viable CD14/TLR4/MD-2 oligomers at the plasma membrane surface, and observe lipopolysaccharide exchange between CD14 and TLR4/MD-2. Collectively, this work helps to unravel the key structural determinants governing endotoxin recognition in the TLR4 innate immune pathway.

On the Calculation of Acyl Chain Order Parameters from Lipid Simulations.

For molecular dynamics simulations of biological membrane systems to live up to the potential of providing accurate atomic level detail into membrane properties and functions, it is essential that the force fields used to model such systems are as accurate as possible. One membrane property that is often used to assess force field accuracy is the carbon-hydrogen (or carbon-deuterium) order parameters of the lipid tails, which can be accurately measured using experimental NMR techniques. There are a variety of analysis tools available to calculate these order parameters from simulations and it is essential that these computational tools work correctly to ensure the accurate assessment of the simulation force fields. In this work we compare many of these computational tools for calculating the order parameters of POPC membranes. While tools that work on all-atom systems and tools that work on saturated lipid tails in general work extremely well, we demonstrate that the majority of the tested tools that calculate the order parameters for unsaturated united-atom lipid tails do so incorrectly. We identify tools that do perform accurate calculations and include one such program with this work, enabling rapid and accurate calculation of united-atom lipid order parameters. Furthermore, we discuss cases in which it is nontrivial to appropriately predict the unsaturated carbon order parameters in united-atom systems. Finally, we examine order parameter splitting for carbon 2 in sn-2 lipid chains, demonstrating substantial deviations from experimental values in several all-atom and united-atom lipid force fields.