PEP-FOLD4: a pH-dependent force field for peptide structure prediction in aqueous solution.

Accurate and fast structure prediction of peptides of less 40 amino acids in aqueous solution has many biological applications, but their conformations are pH- and salt concentration-dependent. In this work, we present PEP-FOLD4 which goes one step beyond many machine-learning approaches, such as AlphaFold2, TrRosetta and RaptorX. Adding the Debye-Hueckel formalism for charged-charged side chain interactions to a Mie formalism for all intramolecular (backbone and side chain) interactions, PEP-FOLD4, based on a coarse-grained representation of the peptides, performs as well as machine-learning methods on well-structured peptides, but displays significant improvements for poly-charged peptides. PEP-FOLD4 is available at http://bioserv.rpbs.univ-paris-diderot.fr/services/PEP-FOLD4. This server is free and there is no login requirement.

Metastable alpha-rich and beta-rich conformations of small Aß42 peptide oligomers.

Probing the structures of amyloid-ß (Aß) peptides in the early steps of aggregation is extremely difficult experimentally and computationally. Yet, this knowledge is extremely important as small oligomers are the most toxic species. Experiments and simulations on Aß42 monomer point to random coil conformations with either transient helical or ß-strand content. Our current conformational description of small Aß42 oligomers is funneled toward amorphous aggregates with some ß-sheet content and rare high energy states with well-ordered assemblies of ß-sheets. In this study, we emphasize another view based on metastable α-helix bundle oligomers spanning the C-terminal residues, which are predicted by the machine-learning AlphaFold2 method and supported indirectly by low-resolution experimental data on many amyloid polypeptides. This finding has consequences in developing novel chemical tools and to design potential therapies to reduce aggregation and toxicity.

Structure and elasticity of healthy and Alzheimer's disease cell membranes revealed by molecular dynamics simulations.

Interactions of amyloid-ß (Aß) peptides with neuronal membrane are associated with the development of Alzheimer's disease (AD). Ganglioside monosialotetrahexosylganglioside (GM1) lipids have been shown to form clusters that induce the structural conversion of Aß and promote the incorporation of Aß into the membrane via the membrane surface electrical potential. Prior to the onset of AD symptoms, GM1 clusters may not have formed but the concentration of GM1 may have already changed, and our question is whether this early concentration modification affects the structure and mechanical properties of the membrane. Using one model for healthy cell membranes and three models for AD cell membranes, we carry out 2 µs all-atom molecular dynamics simulations for each model to compare the structure and elasticity of the two membrane types. The simulations show that at the physiological concentration, 1%-3%, GM1 does not form clusters. The reduction of the GM1 lipid does not significantly alter the area per lipid, the membrane thickness, and the lipid order parameters of the AD membranes. However, the dipole potential, the bending, and twist moduli are decreased for the AD membranes. We suggest that these changes in the AD membranes are factors that could trigger the interaction and incorporation of Aß to the membranes. Finally, we show that changes in the sphingomyelin lipid concentrations do not affect the membrane structure and elasticity.

Evolution of large Aß16-22 aggregates at atomic details and potential of mean force associated to peptide unbinding and fragmentation events.

Atomic characterization of large nonfibrillar aggregates of amyloid polypeptides cannot be determined by experimental means. Starting from ß-rich aggregates of Y and elongated topologies predicted by coarse-grained simulations and consisting of more than 100 Aß16-22 peptides, we performed atomistic molecular dynamics (MD), replica exchange with solute scaling (REST2), and umbrella sampling simulations using the CHARMM36m force field in explicit solvent. Here, we explored the dynamics within 3 µs, the free energy landscape, and the potential of mean force associated with either the unbinding of one single peptide in different configurations within the aggregate or fragmentation events of a large number of peptides. Within the time scale of MD and REST2, we find that the aggregates experience slow global conformational plasticity, and remain essentially random coil though we observe slow beta-strand structuring with a dominance of antiparallel beta-sheets over parallel beta-sheets. Enhanced REST2 simulation is able to capture fragmentation events, and the free energy of fragmentation of a large block of peptides is found to be similar to the free energy associated with fibril depolymerization by one chain for longer Aß sequences.

Amyloid Oligomers: A Joint Experimental/Computational Perspective on Alzheimer's Disease, Parkinson's Disease, Type II Diabetes, and Amyotrophic Lateral Sclerosis.

Protein misfolding and aggregation is observed in many amyloidogenic diseases affecting either the central nervous system or a variety of peripheral tissues. Structural and dynamic characterization of all species along the pathways from monomers to fibrils is challenging by experimental and computational means because they involve intrinsically disordered proteins in most diseases. Yet understanding how amyloid species become toxic is the challenge in developing a treatment for these diseases. Here we review what computer, in vitro, in vivo, and pharmacological experiments tell us about the accumulation and deposition of the oligomers of the (Aß, tau), α-synuclein, IAPP, and superoxide dismutase 1 proteins, which have been the mainstream concept underlying Alzheimer's disease (AD), Parkinson's disease (PD), type II diabetes (T2D), and amyotrophic lateral sclerosis (ALS) research, respectively, for many years.

Multistep molecular mechanisms of Aß16-22 fibril formation revealed by lattice Monte Carlo simulations.

As a model of self-assembly from disordered monomers to fibrils, the amyloid-ß fragment Aß16-22 was subject to past numerous experimental and computational studies. Because dynamics information between milliseconds and seconds cannot be assessed by both studies, we lack a full understanding of its oligomerization. Lattice simulations are particularly well suited to capture pathways to fibrils. In this study, we explored the aggregation of 10 Aß16-22 peptides using 65 lattice Monte Carlo simulations, each simulation consisting of 3 × 109 steps. Based on a total of 24 and 41 simulations that converge and do not converge to the fibril state, respectively, we are able to reveal the diversity of the pathways leading to fibril structure and the conformational traps slowing down the fibril formation.

Recent Computational Advances Regarding Amyloid-ß and Tau Membrane Interactions in Alzheimer's Disease.

The interactions of amyloid proteins with membranes have been subject to many experimental and computational studies, as these interactions contribute in part to neurodegenerative diseases. In this review, we report on recent simulations that have focused on the adsorption and insertion modes of amyloid-ß and tau proteins in membranes. The atomistic-resolution characterization of the conformational changes of these amyloid proteins upon lipid cell membrane and free lipid interactions is of interest to rationally design drugs targeting transient oligomers in Alzheimer's disease.

Elastic moduli of normal and cancer cell membranes revealed by molecular dynamics simulations.

Recent studies indicate that there are mechanical differences between normal cells and cancer cells. Because the cell membrane takes part in a variety of vital processes, we test the hypothesis of whether or not two fundamental alterations in the cell membrane, i.e., the overexpression of phosphatidylserine lipids in the outer leaflet and a reduction in cholesterol concentration, could cause the softening in cancer cells. Adopting ten models of normal and cancer cell membranes, we carry out 1 µs all-atom molecular dynamics simulations to compare the structural properties and elasticity properties of two membrane types. We find that the overexpression of the phosphatidylserine lipids in the outer leaflet does not significantly alter the area per lipid, the membrane thickness, the lipid order parameters and the elasticity moduli of the cancer membranes. However, a reduction in the cholesterol concentration leads to clear changes in those quantities, especially decreases in the bending, tilt and twist moduli. This implies that the reduction of cholesterol concentration in the cancer membranes could contribute to the softening of cancer cells.

Impact of A2T and D23N mutations on C99 homodimer conformations.

The proteolytic cleavage of C99 by γ-secretase is the last step in the production of amyloid-ß (Aß) peptides. Previous studies have shown that membrane lipid composition, cholesterol concentration, and mutation in the transmembrane helix modified the structures and fluctuations of C99. In this study, we performed atomistic molecular dynamics simulations of the homodimer of the 55-residue congener of the C-terminal domain of the amyloid protein precursor, C99(1-55), in a 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine-cholesterol lipid bilayer and compared the conformational ensemble of wild-type (WT) sequence to those of the A2T and D23N variants. These mutations are particularly interesting as the protective Alzheimer's disease (AD) A2T mutation is known to decrease Aß production, whereas the early onset AD D23N mutation does not affect Aß production. We found noticeable differences in the structural ensembles of the three sequences. In particular, A2T varies from both WT and D23N by having long-range effects on the population of the extracellular juxtamembrane helix, the interface between the G29xxx-G33xxx-G37 motifs, and the fluctuations of the transmembrane helical topologies.

Molecular dynamics simulation of cancer cell membrane perforated by shockwave induced bubble collapse.

It has been widely accepted that cancer cells are softer than their normal counterparts. This motivates us to propose, as a proof-of-concept, a method for the efficient delivery of therapeutic agents into cancer cells, while normal cells are less affected. The basic idea of this method is to use a water jet generated by the collapse of the bubble under shockwaves to perforate pores in the cell membrane. Given a combination of shockwave and bubble parameters, the cancer membrane is more susceptible to bending, stretching, and perforating than the normal membrane because the bending modulus of the cancer cell membrane is smaller than that of the normal cell membrane. Therefore, the therapeutic agent delivery into cancer cells is easier than in normal cells. Adopting two well-studied models of the normal and cancer membranes, we perform shockwave induced bubble collapse molecular dynamics simulations to investigate the difference in the response of two membranes over a range of shockwave impulse 15-30 mPa s and bubble diameter 4-10 nm. The simulation shows that the presence of bubbles is essential for generating a water jet, which is required for perforation; otherwise, pores are not formed. Given a set of shockwave impulse and bubble parameters, the pore area in the cancer membrane is always larger than that in the normal membrane. However, a too strong shockwave and/or too large bubble results in too fast disruption of membranes, and pore areas are similar between two membrane types. The pore closure time in the cancer membrane is slower than that in the normal membrane. The implications of our results for applications in real cells are discussed in some details. Our simulation may be useful for encouraging future experimental work on novel approaches for cancer treatment.

Effect of Cholesterol Molecules on Aß1-42 Wild-Type and Mutants Trimers.

Alzheimer's disease displays aggregates of the amyloid-beta (Aß) peptide in the brain, and there is increasing evidence that cholesterol may contribute to the pathogenesis of the disease. Though many experimental and theoretical studies have focused on the interactions of Aß oligomers with membrane models containing cholesterol, an understanding of the effect of free cholesterol on small Aß42 oligomers is not fully established. To address this question, we report on replica exchange with a solute tempering simulation of an Aß42 trimer with cholesterol and compare it with a previous replica exchange molecular dynamics simulation. We show that the binding hot spots of cholesterol are rather complex, involving hydrophobic residues L17-F20 and L30-M35 with a non-negligible contribution of loop residues D22-K28 and N-terminus residues. We also examine the effects of cholesterol on the trimers of the disease-causing A21G and disease-protective A2T mutations by molecular dynamics simulations. We show that these two mutations moderately impact cholesterol-binding modes. In our REST2 simulations, we find that cholesterol is rarely inserted into aggregates but rather attached as dimers and trimers at the surface of Aß42 oligomers. We propose that cholesterol acts as a glue to speed up the formation of larger aggregates; this provides a mechanistic link between cholesterol and Alzheimer's disease.

Molecular Mechanism of Ultrasound-Induced Structural Defects in Liposomes: A Nonequilibrium Molecular Dynamics Simulation Study.

The use of ultrasound in combination with liposomes is a promising approach to improve drug delivery. To achieve an optimal drug release rate, it is important to understand how ultrasound induces pathways on the liposome surface where drugs can be released from the liposome. To this end, we carry out large-scale ultrasound-induced molecular dynamics simulations for three single lipid component liposomes formed from the commonly used phospholipids: 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dipalmitoylphosphatidylcholine (DPPC), or phosphatidylcholine (POPC). The results show that ultrasound induces the detachment of two leaflets of the DOPC surface, suggesting that the drug release pathway may be through the low lipid packing areas on the stretched surface. In contrast, ultrasound induces pore formation on the surface of DPPC and DOPC, where drugs could escape from the liposomes. While the leaflet detachment and transient pore formation are the mechanisms of DOPC and DPPC, respectively, in both liquid-ordered and liquid-disordered phases, the leaflet detachment mechanism is switched to the transient pore formation mechanism on going from the liquid-ordered phase to the liquid-disordered phase in the POPC liposome. By adding 30% mol cholesterol, the leaflet detachment mechanism is observed in all liposomes. We found that the molecular origin that determines a mechanism is the competition between the intraleaflet and interleaflet interacting energy of lipids. The connection to experimental and theoretical modeling is discussed in some detail.

Impact of the Astaxanthin, Betanin, and EGCG Compounds on Small Oligomers of Amyloid Aß40 Peptide.

There is experimental evidence that the astaxanthin, betanin, and epigallocatechin-3-gallate (EGCG) compounds slow down the aggregation kinetics and the toxicity of the amyloid-ß (Aß) peptide. How these inhibitors affect the self-assembly at the atomic level remains elusive. To address this issue, we have performed for each ligand atomistic replica exchange molecular dynamic (REMD) simulations in an explicit solvent of the Aß11-40 trimer from the U-shape conformation and MD simulations starting from Aß1-40 dimer and tetramer structures characterized by different intra- and interpeptide conformations. We find that the three ligands have similar binding free energies on small Aß40 oligomers but very distinct transient binding sites that will affect the aggregation of larger assemblies and fibril elongation of the Aß40 peptide.

Molecular mechanism of ultrasound interaction with a blood brain barrier model.

The brain is strictly protected by the blood brain barrier preventing the crossing of therapeutics to treat brain diseases. The high and low intensity focused ultrasound methods have been used to temporarily open the blood brain barrier, facilitating the transport of drugs. The methods are very promising because the opening is transient, localized, and noninvasive. However, the molecular mechanism of the opening is unknown, and this limits the development and application of these methods. With this in mind, we carry out a molecular dynamics simulation study to understand the interaction of ultrasound with the cell membrane and the tight junction. Our minimal blood brain barrier model is composed of two lipid bilayers, mimicking two portions of neighboring cells, connected together by a tight junction formed by a pair of two cis-dimers of the claudin-5 protein. Using an experimental ultrasound frequency of 50 MHz, simulations show that at low intensities, ultrasound does not impact the structure of the cell membranes and tight junction, implying that the direct interaction of ultrasound with the blood brain barrier is not responsible for the experimentally observed opening. At high intensities, the ultrasound pulls the monolayers of individual cell membrane lipid bilayers apart, creating air compartments inside the bilayers. This reduces the free energy barrier for the translocation of drugs across the lipid bilayer and enhances drug permeability. At very high intensities, the two monolayers are largely separated, resulting in cell damage and implying that the blood brain barrier is primarily opened at the experimentally observed damaged areas.

Probing the quality control mechanism of the Escherichia coli twin-arginine translocase with folding variants of a de novo-designed heme protein.

Protein transport across the cytoplasmic membrane of bacterial cells is mediated by either the general secretion (Sec) system or the twin-arginine translocase (Tat). The Tat machinery exports folded and cofactor-containing proteins from the cytoplasm to the periplasm by using the transmembrane proton motive force as a source of energy. The Tat apparatus apparently senses the folded state of its protein substrates, a quality-control mechanism that prevents premature export of nascent unfolded or misfolded polypeptides, but its mechanistic basis has not yet been determined. Here, we investigated the innate ability of the model Escherichia coli Tat system to recognize and translocate de novo-designed protein substrates with experimentally determined differences in the extent of folding. Water-soluble, four-helix bundle maquette proteins were engineered to bind two, one, or no heme b cofactors, resulting in a concomitant reduction in the extent of their folding, assessed with temperature-dependent CD spectroscopy and one-dimensional 1H NMR spectroscopy. Fusion of the archetypal N-terminal Tat signal peptide of the E. coli trimethylamine-N-oxide (TMAO) reductase (TorA) to the N terminus of the protein maquettes was sufficient for the Tat system to recognize them as substrates. The clear correlation between the level of Tat-dependent export and the degree of heme b-induced folding of the maquette protein suggested that the membrane-bound Tat machinery can sense the extent of folding and conformational flexibility of its substrates. We propose that these artificial proteins are ideal substrates for future investigations of the Tat system's quality-control mechanism.

VLITL is a major cross-ß-sheet signal for fibrinogen Aα-chain frameshift variants.

The first case of hereditary fibrinogen Aα-chain amyloidosis was recognized >20 years ago, but disease mechanisms still remain unknown. Here we report detailed clinical and proteomics studies of a French kindred with a novel amyloidogenic fibrinogen Aα-chain frameshift variant, Phe521Leufs, causing a severe familial form of renal amyloidosis. Next, we focused our investigations to elucidate the molecular basis that render this Aα-chain variant amyloidogenic. We show that a 49-mer peptide derived from the C-terminal part of the Phe521Leufs chain is deposited as fibrils in the patient's kidneys, establishing that only a small portion of Phe521Leufs directly contributes to amyloid formation in vivo. In silico analysis indicated that this 49-mer Aα-chain peptide contained a motif (VLITL), with a high intrinsic propensity for ß-aggregation at residues 44 to 48 of human renal fibrils. To experimentally verify the amyloid propensity of VLITL, we generated synthetic Phe521Leufs-derived peptides and compared their capacity for fibril formation in vitro with that of their VLITL-deleted counterparts. We show that VLITL forms typical amyloid fibrils in vitro and is a major signal for cross-ß-sheet self-association of the 49-mer Phe521Leufs peptide identified in vivo, whereas its absence abrogates fibril formation. This study provides compelling evidence that VLITL confers amyloidogenic properties to Aα-chain frameshift variants, yielding a previously unknown molecular basis for the pathogenesis of Aα-chain amyloidosis.

Nonequilibrium atomistic molecular dynamics simulation of tubular nanomotor propelled by bubble propulsion.

We develop a molecular nanoscaled model for tubular motors propelled by bubble propulsion. The motor is modeled by a carbon nanotube, and the bubble is represented by a particle interacting with water by a time-dependent potential. Effects of liquid viscosity, fuel concentration, geometry, and size of the tube on the performance of the motor are effectively encoded into two parameters: time scales of the bubble expansion and bubble formation. Our results are qualitatively consistent with experimental data of much larger motors. Simulations suggest that (i) the displacement of the tube is optimized if two time scales are as short as possible, (ii) the compromise between the performance and fuel consumption is achieved if the bubble formation time is shorter than the velocity correlation time of the tube, (iii) the motor efficiency is higher with slow expansion, short formation of the bubble than fast growth but long formation time, and (iv) the tube is propelled by strong forces on the order of mN, reaching high speeds up to ∼60 m/s. Our simulation may be useful for refining and encouraging future experimental work on nanomotors having the size of a few nanometers. The tiny size and high speed motors could have great potential applications in real life.

Interaction mechanism between the focused ultrasound and lipid membrane at the molecular level.

Focused ultrasound (FUS) has a wide range of medical applications. Nowadays, the diagnostic and therapeutic ultrasound procedures are routinely used; effects of ultrasound on biological systems at the molecular level are, however, not fully understood. Experimental results on the interaction of the cell membrane, a simplest but important system component, with ultrasound are controversial. Molecular dynamics (MD) simulations could provide valuable insights, but there is no single study on the mechanism of the FUS induced structural changes in cell membranes. With this in mind, we develop a simple method to include FUS into a standard MD simulation. Adopting the 1,2-dioleoyl-sn-glycero-3-phosphocholine lipid membrane as a representative model described by the MARTINI coarse-grained force field, and using experimental values of the ultrasound frequency and intensity, we show that the heat and bubble cavitation are not the primary direct mechanisms that cause structural changes in the membrane. The spatial pressure gradients between the focused and free regions and between the parallel and perpendicular directions to the membrane are the origin of the mechanism. These gradients force lipids to move out of the focused region, forming a lipid flow along the membrane diagonal. Lipids in the free region move in the opposite direction due to the conservation of the total momentum. These opposite motions create wrinkles along the membrane diagonal at low FUS intensities and tear up the membrane at high FUS intensities. Once the membrane is torn up, it is not easy to reform. The implication of our findings in the FUS-induced drug delivery is discussed in some detail.

Protein-RNA complexation driven by the charge regulation mechanism.

Electrostatic interactions play a pivotal role in many (bio)molecular association processes. The molecular organization and function in biological systems are largely determined by these interactions from pure Coulombic contributions to more peculiar mesoscopic forces due to ion-ion correlation and proton fluctuations. The latter is a general electrostatic mechanism that gives attraction particularly at low electrolyte concentrations. This charge regulation mechanism due to titrating amino acid and nucleotides residues is discussed here in a purely electrostatic framework. By means of constant-pH Monte Carlo simulations based on a fast coarse-grained titration proton scheme, a new computer molecular model was devised to study protein-RNA interactions. The complexation between the RNA silencing suppressor p19 viral protein and the 19-bp small interfering RNA was investigated at different solution pH and salt conditions. The outcomes illustrate the importance of the charge regulation mechanism that enhances the association between these macromolecules in a similar way as observed for other protein-polyelectrolyte systems typically found in colloidal science. Due to the highly negative charge of RNA, the effect is more pronounced in this system as predicted by the Kirkwood-Shumaker theory. Our results contribute to the general physico-chemical understanding of macromolecular complexation and shed light on the extensive role of RNA in the cell's life.

Multi-scale simulations of biological systems using the OPEP coarse-grained model.

Biomolecules are complex machines that are optimized by evolution to properly fulfill or contribute to a variety of biochemical tasks in the cellular environment. Computer simulations based on quantum mechanics and atomistic force fields have been proven to be a powerful microscope for obtaining valuable insights into many biological, physical, and chemical processes. Many interesting phenomena involve, however, a time scale and a number of degrees of freedom, notably if crowding is considered, that cannot be explored at an atomistic resolution. To bridge the gap between reality and simulation, many different advanced computational techniques and coarse-grained (CG) models have been developed. Here, we report some applications of the CG OPEP protein model to amyloid fibril formation, the response of catch-bond proteins to two types of fluid flow, and interactive simulations to fold peptides with well-defined 3D structures or with intrinsic disorder.