Simulations of Protein Aggregation: Insights from Atomistic and Coarse-Grained Models

Biophysical Principles Emerging from Experiments on Protein-Protein Association and Aggregation

Protein-protein association and aggregation are fundamental processes that play critical roles in various biological phenomena, from cellular signaling to disease progression. Understanding the underlying biophysical principles governing these processes is crucial for elucidating their mechanisms and developing strategies for therapeutic intervention. In this review, we provide an overview of recent experimental studies focused on protein-protein association and aggregation. We explore the key biophysical factors that influence these processes, including protein structure, conformational dynamics, and intermolecular interactions. We discuss the effects of environmental conditions such as temperature, pH and related buffer-specific effects, and ionic strength and related ion-specific effects on protein aggregation. The effects of polymer crowders and sugars are also addressed. We list the techniques used to study aggregation. We analyze emerging trends and challenges in the field, including the development of computational models and the integration of multidisciplinary approaches for a comprehensive understanding of protein-protein association and aggregation.

Experimental and molecular dynamics simulation studies on the physical properties of three HBc-VLP derivatives as nanoparticle protein vaccine candidates

Self-assembling virus-like particles (VLPs) are promising platforms for vaccine development. However, the unpredictability of the physical properties, such as self-assembly capability, hydrophobicity, and overall stability in engineered protein particles fused with antigens, presents substantial challenges in their downstream processing. We envision that these challenges can be addressed by combining more precise computer-aided molecular dynamics (MD) simulations with experimental studies on the modified products, with more to-date forcefield descriptions and larger models closely resembling real assemblies, realized by rapid advancement in computing technology. In this study, three chimeric designs based on the hepatitis B core (HBc) protein as model vaccine candidates were constructed to study and compare the influence of inserted epitopes as well as insertion strategy on HBc modifications. Large partial VLP models containing 17 chains for the HBc chimeric model vaccines were constructed based on the wild-type (wt) HBc assembly template. The findings from our simulation analysis have demonstrated good consistency with experimental results, pertaining to the surface hydrophobicity and overall stability of the chimeric vaccine candidates. Furthermore, the different impact of foreign antigen insertions on the HBc scaffold was investigated through simulations. It was found that separately inserting two epitopes into the HBc platform at the N-terminal and the major immunogenic regions (MIR) yields better results compared to a serial insertion at MIR in terms of protein structural stability. This study substantiates that an MD-guided design approach can facilitate vaccine development and improve its manufacturing efficiency by predicting products with extreme surface hydrophobicity or structural instability.

Molecular Mechanism of Interaction between DNA Aptamer and Receptor-Binding Domain of Severe Acute Respiratory Syndrome Coronavirus 2 Variants Revealed by Steered Molecular Dynamics Simulations

The ongoing SARS-CoV-2 pandemic has underscored the urgent need for versatile and rapidly deployable antiviral strategies. While vaccines have been pivotal in controlling the spread of the virus, the emergence of new variants continues to pose significant challenges to global health. Here, our study focuses on a novel approach to antiviral therapy using DNA aptamers, short oligonucleotides with high specificity and affinity for their targets, as potential inhibitors against the spike protein of SARS-CoV-2 variants Omicron and JN.1. Our research utilizes steered molecular dynamics (SMD) simulations to elucidate the binding mechanisms of a specifically designed DNA aptamer, AM032-4, to the receptor-binding domain (RBD) of the aforementioned variants. The simulations reveal detailed molecular insights into the aptamer-RBD interaction, demonstrating the aptamer's potential to maintain effective binding in the face of rapid viral evolution. Our work not only demonstrates the dynamic interaction between aptamer-RBD for possible antiviral therapy but also introduces a computational method to study aptamer-protein interactions.

Molecular Driving Forces in the Self-Association of Silaffin Peptide R5 from MD Simulations

The 19-residue silaffin-R5 peptide has been widely studied for its ability to precipitate uniform SiO2 particles through mild temperature and pH pathways, in the absence of any organic solvents. There is consensus that post-translational modification (PTM) of side chains has a large impact on the biomineralization process. Thus, it is imperative to understand the precise mechanisms that dictate the formation of SiO2 from R5 peptide, including the effects of PTM on peptide aggregation and peptide-surface adsorption. In this work, we use molecular dynamics (MD) simulations to study the aggregation of R5 dimer with multiple PTMs, with the presence of different ions in solution. Since this system has strong interactions with deep metastable states, we use parallel bias metadynamics with partitioned families to efficiently sample the different states of the system. We find that peptide aggregation is a prerequisite for biomineralization. We observe that the electrostatic interactions are essential in the R5 dimer aggregation; for wild type R5 that only has positively charged residues, phosphate ions HPO4 2- in the solution form a bridge between two peptides and are essential for peptide aggregation.

Multiscale Simulations to Discover Self-Assembled Oligopeptides: A Benchmarking Study

Peptide self-assembly is critical for biomedical and material discovery and production. While it is costly to experimentally test every possible peptide design, computational assessment provides an affordable solution to evaluate many designs and prioritize synthesis and characterization. Following a theoretical investigation, we present a systematic analysis of all-atom and coarse-grained simulations to predict peptide self-assembly. Benchmarking studies of two model dipeptides allow us to assess the impacts of intrinsic properties (such as amino acids and terminal modifications) and external environment (such as salinity) on the simulated aggregation. Further examination of 20 oligopeptides containing two to five amino acids shows good agreement among our theory, simulations, and prior experimental observations. The success rate of our prediction is 90%. Therefore, our theory, simulation, and analysis can be useful to identify peptide designs that can self-assemble and predict the potential nanostructures. These findings lay the ground for future virtual screening of peptide-assembled nanostructures and computer-aided biologics design.

Protein Association in Solution: Statistical Mechanical Modeling

Protein molecules associate in solution, often in clusters beyond pairwise, leading to liquid phase separations and high viscosities. It is often impractical to study these multi-protein systems by atomistic computer simulations, particularly in multi-component solvents. Instead, their forces and states can be studied by liquid state statistical mechanics. However, past such approaches, such as the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, were limited to modeling proteins as spheres, and contained no microscopic structure-property relations. Recently, this limitation has been partly overcome by bringing the powerful Wertheim theory of associating molecules to bear on protein association equilibria. Here, we review these developments.

Nucleation and Growth of Amyloid Fibrils

The formation of amyloid fibrils is a complex phenomenon that remains poorly understood at the atomic scale. Herein, we perform extended unbiased all-atom simulations in explicit solvent of a short amphipathic peptide to shed light on the three mechanisms accounting for fibril formation, namely, nucleation via primary and secondary mechanisms, and fibril growth. We find that primary nucleation takes place via the formation of an intermediate state made of two laminated β-sheets oriented perpendicular to each other. The amyloid fibril spine subsequently emerges from the rotation of these β-sheets to account for peptides that are parallel to each other and perpendicular to the main axis of the fibril. Growth of this spine, in turn, takes place via a dock-and-lock mechanism. We find that peptides dock onto the fibril tip either from bulk solution or after diffusing on the fibril surface. The latter docking pathway contributes significantly to populate the fibril tip with peptides. We also find that side chain interactions drive the motion of peptides in the lock phase during growth, enabling them to adopt the structure imposed by the fibril tip with atomic fidelity. Conversely, the docked peptide becomes trapped in a local free energy minimum when docked-conformations are sampled randomly. Our simulations also highlight the role played by nonpolar fibril surface patches in catalyzing and orienting the formation of small cross-β structures. More broadly, our simulations provide important new insights into the pathways and interactions accounting for primary and secondary nucleation as well as the growth of amyloid fibrils.

Aggregation and partitioning of amyloid peptide fragments in the presence of a lipid bilayer: A coarse grained molecular dynamics study

Amyloidogenicity and toxicity of amyloid peptides have been linked to the peptide aggregation and interactions with lipid bilayers. In this work we used the coarse grained MARTINI model to study the aggregation and partitioning of amyloid peptide fragments Aβ(1-28) and Aβ(25-35) in the presence of a dipalmitoylphosphatidylcholine bilayer. We explored the peptide aggregation starting from three initial spatial arrangements where free monomers were placed in solution outside the membrane, at the membrane-solution interface, or in the membrane. We found that Aβ(1-28) and Aβ(25-35) interact with the bilayer quite differently. The Aβ(1-28) fragments show strong peptide-peptide and peptide-lipid interactions leading to irreversible aggregation where the aggregates stay confined to their initial spatial location. The Aβ(25-35) fragments show weaker peptide-peptide and peptide-lipid interaction leading to reversible aggregation and accumulation at the membrane-solution interface irrespective of their initial spatial arrangement. Those findings can be explained in terms of the shape of the potential of mean force for the single-peptide translocation across the membrane.

Aggregation of an Amyloidogenic Peptide on Gold Surfaces

Solid surfaces have been shown to affect the aggregation and assembly of many biomolecular systems. One important example is the formation of protein fibrils, which can occur on a range of biological and synthetic surfaces. The rate of fibrillation depends on both the protein structure and the surface chemistry, with the different molecular and oligomer structures adopted by proteins on surfaces likely to be crucial. In this paper, the aggregation of the model amyloidogenic peptide, Aβ(16-22), corresponding to a hydrophobic segment of the amyloid beta protein on a gold surface is studied using molecular dynamics simulation. Previous simulations of this peptide on gold surfaces have shown that it adopts conformations on surfaces that are quite different from those in bulk solution. These simulations show that this then leads to significant differences in the oligomer structures formed in solution and on gold surfaces. In particular, oligomers formed on the surface are low in beta-strands so are unlike the structures formed in bulk solution. When oligomers formed in solution adsorb onto gold surfaces they can then restructure themselves. This can then help explain the inhibition of Aβ(16-22) fibrillation by gold surfaces and nanoparticles seen experimentally.

Modified Smoluchowski Rate Equations for Aggregation and Fragmentation in Finite Systems

Protein self-assembly into supramolecular structures is important for cell biology. Theoretical methods employed to investigate protein aggregation and analogous processes include molecular dynamics simulations, stochastic models, and deterministic rate equations based on the mass-action law. In molecular dynamics simulations, the computation cost limits the system size, simulation length, and number of simulation repeats. Therefore, it is of practical interest to develop new methods for the kinetic analysis of simulations. In this work we consider the Smoluchowski rate equations modified to account for reversible aggregation in finite systems. We present several examples and argue that the modified Smoluchowski equations combined with Monte Carlo simulations of the corresponding master equation provide an effective tool for developing kinetic models of peptide aggregation in molecular dynamics simulations.

Governing dynamics and preferential binding of the AXH domain influence the aggregation pathway of Ataxin-1

The present state of understanding the mechanism of Spinocerebellar Ataxia-1, a fatal neurodegenerative disease linked to the protein Ataxin-1 (ATXN1), is baffled by a set of self-contradictory, and hence, inconclusive observations. This fallacy poses a bottleneck to the effective designing of curable drugs as the field is currently missing the specific druggable site. To understand the fundamentals of pathogenesis, we tried to decipher the intricacies of the extremely complicated landscape by targeting the relevant species that supposedly dictate the structure-function paradigm. The atomic-level description and characterization of the dynamism of the systems reveal the existence of structural polymorphism in all the leading stakeholders of the overall system. The very existence of conformational heterogeneity in every species creates numerous possible combinations of favorable interactions because of the variability in segmental cross-talks and hence claims its role in the choice of routes between functional activity and dysfunctional disease-causing aggregation. Despite this emergent configurational diversity, there is a common mode of operative intermolecular forces that dictates the extent of stability of all the multimeric complexes due to the localized population of a specific type of residue. The present research proposes a dynamic switch mechanism between aggregability and functional activity, based on the logical interpretation of the estimated variables, which is practically dictated by the effective concentration of the interacting species involved in the cell.

Zwitterionic Biomaterials

The term "zwitterionic polymers" refers to polymers that bear a pair of oppositely charged groups in their repeating units. When these oppositely charged groups are equally distributed at the molecular level, the molecules exhibit an overall neutral charge with a strong hydration effect via ionic solvation. The strong hydration effect constitutes the foundation of a series of exceptional properties of zwitterionic materials, including resistance to protein adsorption, lubrication at interfaces, promotion of protein stabilities, antifreezing in solutions, etc. As a result, zwitterionic materials have drawn great attention in biomedical and engineering applications in recent years. In this review, we give a comprehensive and panoramic overview of zwitterionic materials, covering the fundamentals of hydration and nonfouling behaviors, different types of zwitterionic surfaces and polymers, and their biomedical applications.

Inhibition of Amyloid Nucleation by Steric Hindrance

Despite recent experiments and simulations suggesting that small-molecule inhibitors and some post-translational modifications (e.g., glycosylation and ubiquitination) can suppress the pathogenic aggregation of proteins due to steric hindrance, the effect of steric hindrance on amyloid formation has not been systematically studied. Based on Monte Carlo simulations using a coarse-grained model for amyloidogenic proteins and a hard sphere acting as steric hindrance, we investigated how steric hindrance on proteins could affect amyloid formation, particularly two steps of primary nucleation, namely, oligomerization and conformational conversion into a β-sheet-enriched nucleus. We found that steric spheres played an inhibitory role in oligomerization with the effect proportional to the sphere radius R_S, which we attributed to the decline in the nonspecific attractions between proteins. During the second step, small steric spheres facilitated the conformational conversion of proteins while large ones suppressed the conversion. The overall steric effect on amyloid nucleation was inhibitory regardless of R_S. As R_S increased, oligomeric assemblies changed from amorphous into sheet-like, structurally ordered species, reminiscent of the structure of amyloid fibrils. The oligomers with large R_S were off-pathway with their ordered structures induced by the competition between steric hindrance and nonspecific attractions of soluble proteins. Interestingly, the equimolar mixture of proteins with and without steric hindrance amplified the sterically inhibitory effect by increasing the energy barrier of protein's conformational conversion. The physical mechanisms and biological implications of the above results are discussed. Our findings improve the current understanding of how nature regulates protein aggregation and amyloid formation by steric hindrance.

Atomistic Pictures of Self-Assembled Helical Peptide Nanofibers

Spontaneous self-assembly of peptides has been at the forefront of supramolecular chemistry and materials science research over the last two decades. Despite the wealth of information on the morphology of the assembled objects, atomic resolution details of molecular arrangements inside them are largely unknown. In this paper, we investigated non-covalent assemblies of zwitterionic l-phenylalanine tripeptides in water using all-atom explicit-solvent molecular dynamics computer simulations. Our studies produced atomistic pictures of spontaneously assembled nanofibers composed of hundreds of peptide molecules. The dimensions of the nanofibers varied from 10 to 18 nm, with irregular helical twists along the long axes. Previously published experimental data, acquired under similar conditions, provided direct validation of the fibrous morphology and indirect support for the non-trivial helicity observed in our simulations. Quantitative analyses of peptide-water and peptide-peptide interactions revealed heterogeneous local environments of molecules across the nanometer length scales. The combination of electrostatic, hydrogen bonding, van der Waals, and hydrophobic interactions, adopted by a single molecule, was dependent on its relative position inside the fiber. Despite the presence of three hydrophobic phenyl groups, very few molecules were found to be completely shielded from the surrounding water, indicating a subtle role of the hydrophobic effect. Limited conformational flexibility of the tripeptide, along with bare electrostatic interactions, appeared to play a crucial role in the emergence of fibrous morphology of the nanostructures. Our analyses led us to formulate plausible qualitative explanations of the assembly behavior in terms of thermodynamic driving forces and kinetic considerations. We established a clear relationship between details of chemical interactions operating within few molecules and characteristics of the self-assembled states at much longer length scales.

Side Chain Geometry Determines the Fibrillation Propensity of a Minimal Two-Beads-per-Residue Peptide Model

The molecular mechanism of fibrillation is an important issue for understanding peptide aggregation. In our previous work, we demonstrated that the interchain attraction and intrachain bending stiffness control the aggregation kinetics and transient aggregate morphologies of a one-bead-per-residue implicit solvent peptide model. However, that model did not lead to fibrillation. In this work, we study the molecular origin of fibril formation using a two-beads-per-residue model, where one bead represents the backbone residue atoms and the other the side chain atoms. We show that the side chain geometry determines the fibrillation propensity that is further modulated by the modified terminal beads. This allows us to bring out the effects of side chain geometry and terminal capping on the fibrillation propensity. Our model does not assume a secondary structure and is, perhaps, the simplest bead-based chain model leading to fibrillation.

Scan-Find-Scan-Model: Discrete Site-Targeted Suppressor Design Strategy for Amyloid-β

Alzheimer's disease is undoubtedly the most well-studied neurodegenerative disease. Consequently, the amyloid-β (Aβ) protein ranks at the top in terms of getting attention from the scientific community for structural property-based characterization. Even after decades of extensive research, there is existing volatility in terms of understanding and hence the effective tackling procedures against the disease that arises due to the lack of knowledge of both specific target- and site-specific drugs. Here, we develop a multidimensional approach based on the characterization of the common static-dynamic-thermodynamic trait of the monomeric protein, which efficiently identifies a small target sequence that contains an inherent tendency to misfold and consequently aggregate. The robustness of the identification of the target sequence comes with an abundance of a priori knowledge about the length and sequence of the target and hence guides toward effective designing of the target-specific drug with a very low probability of bottleneck and failure. Based on the target sequence information, we further identified a specific mutant that showed the maximum potential to act as a destabilizer of the monomeric protein as well as enormous success as an aggregation suppressor. We eventually tested the drug efficacy by estimating the extent of modulation of binding affinity existing within the fibrillar form of the Aβ protein due to a single-point mutation and hence provided a proof of concept of the entire protocol.

Distribution and Structure Analysis of Fibril-Forming Peptides Focusing on Concentration Dependency

We focus on the concentration dependency of fibril-forming peptides, which have the potential of aggregation by themselves. In this study, we performed replica-exchange molecular dynamics simulations of Lys-Phe-Phe-Glu (KFFE) fragments, which are known to form fibrils in experiments under different concentration environments. The analysis by static structure factors suggested that the density fluctuation of the KFFE fragments becomes large as the concentration increases. It was also found that the number of β-structures and oligomers also increases under a high concentration environment. Hence, a high concentration environment of fibril-forming peptides is likely to cause protein aggregation.

Computational models for studying physical instabilities in high concentration biotherapeutic formulations

Computational prediction of the behavior of concentrated protein solutions is particularly advantageous in early development stages of biotherapeutics when material availability is limited and a large set of formulation conditions needs to be explored. This review provides an overview of the different computational paradigms that have been successfully used in modeling undesirable physical behaviors of protein solutions with a particular emphasis on high-concentration drug formulations. This includes models ranging from all-atom simulations, coarse-grained representations to macro-scale mathematical descriptions used to study physical instability phenomena of protein solutions such as aggregation, elevated viscosity, and phase separation. These models are compared and summarized in the context of the physical processes and their underlying assumptions and limitations. A detailed analysis is also given for identifying protein interaction processes that are explicitly or implicitly considered in the different modeling approaches and particularly their relations to various formulation parameters. Lastly, many of the shortcomings of existing computational models are discussed, providing perspectives and possible directions toward an efficient computational framework for designing effective protein formulations.

Systematic Search for a Predictor for the Clinical Observables of Alzheimer's Disease

One of the prevailing life-threatening incurable neurodegenerative diseases that are presently endangering human society as a whole, and hence, baffling the entire spectrum of the scientific and pharmaceutical world, is Alzheimer's disease (AD). AD is a manifestation of self-assembly of both wild-type (sporadic) and mutated (familial) forms of the amyloid-β peptide, a proteolytic product of the amyloid precursor protein, where the self-assembly results in the genesis of pathogenic fibrillar aggregates. Currently prevailing diagnostic and hence therapeutic challenges originate from the unavailability of a specific predictor for clinical observables. The continuous emergence of novel pathogenic mutants with unpredictable phenotypes adds immensely to the nonspecific nature of the problem. The current research reports a simple physical parameter, the binding affinity of a protofilament to its protofibril, which predicts the clinical observables of familial AD with astounding accuracy and more importantly, without any adjustable parameters.

Aggregation and structure of amyloid β-protein

Alzheimer's disease (AD) is the most common age-related neurodegenerative disorder and is characterized by major pathological hallmarks in the brain, including plaques composed of amyloid β-protein (Aβ) and neurofibrillary tangles of tau protein. Genetic studies, biochemical data, and animal models have suggested that Aβ is a critical species in the pathogenesis of AD. Aβ molecules aggregate to form oligomers, protofibrils (PFs), and mature fibrils. Because of their instability and structural heterogeneity, the misfolding and aggregation of Aβ is a highly complex process, leading to a variety of aggregates with different structures and morphologies. However, the elucidation of Aβ molecules is essential because they are believed to play an important role in AD pathogenesis. Recent combination studies using nuclear magnetic resonance (NMR) and cryo-electron microscopy (cryo-EM) have primarily revealed more detailed information about their aggregation process, including fibril extension and secondary nucleation, and the structural polymorphism of the fibrils under a variety of some conditions, including the actual brain. This review attempts to summarize the existing information on the major properties of the structure and aggregation of Aβ.

Distinct Modes of Action of IAPP Oligomers on Membranes

Islet amyloid polypeptide (IAPP, also known as amylin) is a peptide hormone that is co-secreted with insulin by pancreatic β-cells and forms amyloid aggregates in type II diabetes. Various lines of evidence indicate that oligomers of this peptide may induce toxicity by disrupting or forming pores in cell membranes, but the structure of these pores is unknown. Here, we create models of pores for both helical and β-structured peptides using implicit membrane modeling and test their stability using multimicrosecond all-atom simulations. We find that the helical peptides behave similarly to antimicrobial peptides; they remain stably inserted in a highly tilted or partially unfolded configuration creating a narrow water channel. Parallel helix orientation creates a somewhat larger pore. An octameric β barrel of parallel β-hairpins is highly stable in the membrane, whereas the corresponding barrel made of antiparallel hairpins is not. We propose that certain experiments probe the helical pore state while others probe the β-structured pore state; this provides a possible explanation for lack of correlation that is sometimes observed between in vivo toxicity and in vitro liposome permeabilization experiments.

Diverse Aggregation Kinetics Predicted by a Coarse-Grained Peptide Model

Protein and peptide aggregation is a ubiquitous phenomenon with implications in medicine, pharmaceutical industry, and materials science. An important issue in peptide aggregation is the molecular mechanism of aggregate nucleation and growth. In many experimental studies, sigmoidal kinetics curves show a clear lag phase ascribed to nucleation; however, experimental studies also show downhill kinetics curves, where the monomers decay continuously and no lag phase can be seen. In this work, we study peptide aggregation kinetics using a coarse-grained implicit solvent model introduced in our previous work. Our simulations explore the hypothesis that the interplay between interchain attraction and intrachain bending stiffness controls the aggregation kinetics and transient aggregate morphologies. Indeed, our model reproduces the aggregation modes seen in experiment: no observed aggregation, nucleated aggregation, and rapid downhill aggregation. We find that the interaction strength is the primary parameter determining the aggregation mode, whereas the stiffness is a secondary parameter modulating the transient morphologies and aggregation rates: more attractive and stiff chains aggregate more rapidly and the transient morphologies are more ordered. We also explore the effects of the initial monomer concentration and the chain length. As the concentration decreases, the aggregation mode shifts from downhill to nucleated and no-aggregation. This concentration effect is in line with an experimental observation that the transition between downhill and nucleated kinetics is concentration-dependent. We find that longer peptides can aggregate at conditions where short peptides do not aggregate at all. It supports an experimental observation that the elongation of a homopeptide, e.g., polyglutamine, can increase the aggregation propensity.

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 (Aβ, 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.

Protein aggregation: in silico algorithms and applications

Protein aggregation is a topic of immense interest to the scientific community due to its role in several neurodegenerative diseases/disorders and industrial importance. Several in silico techniques, tools, and algorithms have been developed to predict aggregation in proteins and understand the aggregation mechanisms. This review attempts to provide an essence of the vast developments in in silico approaches, resources available, and future perspectives. It reviews aggregation-related databases, mechanistic models (aggregation-prone region and aggregation propensity prediction), kinetic models (aggregation rate prediction), and molecular dynamics studies related to aggregation. With a multitude of prediction models related to aggregation already available to the scientific community, the field of protein aggregation is rapidly maturing to tackle new applications.

Studying the mechanism of phase separation in aqueous solutions of globular proteins via molecular dynamics computer simulations

Proteins are the most abundant biomacromolecules in living cells, where they perform vital roles in virtually every biological process. To maintain their function, proteins need to remain in a stable (native) state. Inter- and intramolecular interactions in aqueous protein solutions govern the fate of proteins, as they can provoke their unfolding or association into aggregates. The initial steps of protein aggregation are difficult to capture experimentally, therefore we used molecular dynamics simulations in this study. We investigated the initial phase of aggregation of two different lysozymes, hen egg-white (HEWL) and T4 WT* lysozyme and also human lens γ-D crystallin by using atomistic simulations. We monitored the phase stability of their aqueous solutions by calculating time-dependent density fluctuations. We found that all proteins remained in their compact form despite aggregation. With an extensive analysis of intermolecular residue-residue interactions we discovered that arginine is of paramount importance in the initial stage of aggregation of HEWL and γ-D crystallin, meanwhile lysine was found to be the most involved amino acid in forming initial contacts between T4 WT* molecules.

Protein aggregation detection with fluorescent macromolecular and nanostructured probes: challenges and opportunities

Nanoprobes based on various nanomaterials, polymers or AIEgens are overcoming previous limitations for diagnosis and therapy of early-stage protein aggregation.

Temperature dependent aggregation mechanism and pathway of lysozyme: By all atom and coarse grained molecular dynamics simulation

Aggregation of protein causes various diseases including Alzheimer's disease, Parkinson's disease, and type II diabetes. It was found that aggregation of protein depends on many factors like temperature, pH, salt type, salt concentration, ionic strength, protein concentration, co solutes. Here we have tried to capture the aggregation mechanism and pathway of hen egg white lysozyme using molecular dynamics simulations at two different temperatures; 300 K and 340 K. Along with the all atom simulations to get the atomistic details of aggregation mechanism, we have used coarse grained simulation with MARTINI force field to monitor the aggregation for longer duration. Our results suggest that due to the aggregation, changes in the conformation of lysozyme are more at 340 K than at 300 K. The change in the conformation of the lysozyme at 300 K is mainly due to aggregation where at 340 K change in conformation of lysozyme is due to both aggregation and temperature. Also, a more compact aggregated system is formed at 340 K.

Clustering and Fibril Formation during GNNQQNY Aggregation: A Molecular Dynamics Study

The precise kinetic pathways of peptide clustering and fibril formation are not fully understood. Here we study the initial clustering kinetics and transient cluster morphologies during aggregation of the heptapeptide fragment GNNQQNY from the yeast prion protein Sup35. We use a mid-resolution coarse-grained molecular dynamics model of Bereau and Deserno to explore the aggregation pathways from the initial random distribution of free monomers to the formation of large clusters. By increasing the system size to 72 peptides we could follow directly the molecular events leading to the formation of stable fibril-like structures. To quantify those structures we developed a new cluster helicity parameter. We found that the formation of fibril-like structures is a cooperative processes that requires a critical number of monomers, M⋆≈25, in a cluster. The terminal tyrosine residue is the structural determinant in the formation of helical fibril-like structures. This work supports and quantifies the two-step aggregation model where the initially formed amorphous clusters grow and, when they are large enough, rearrange into mature twisted structures. However, in addition to the nucleated fibrillation, growing aggregates undergo further internal reorganization, which leads to more compact structures of large aggregates.

Latent Models of Molecular Dynamics Data: Automatic Order Parameter Generation for Peptide Fibrillization

Variational autoencoders are artificial neural networks with the capability to reduce highly dimensional sets of data to smaller dimensional, latent representations. In this work, these models are applied to molecular dynamics simulations of the self-assembly of coarse-grained peptides to obtain a singled-valued order parameter for amyloid aggregation. This automatically learned order parameter is constructed by time-averaging the latent parametrizations of internal coordinate representations and compared to the nematic order parameter which is commonly used to study ordering of similar systems in literature. It is found that the latent space value provides more tailored insight into the aggregation mechanism's details, correctly identifying fibril formation in instances where the nematic order parameter fails to do so. A means is provided by which the latent space value can be analyzed so that the major contributing internal coordinates are identified, allowing for a direct interpretation of the latent space order parameter in terms of the behavior of the system. The latent model is found to be an effective and convenient way of representing the data from the dynamic ensemble and provides a means of reducing the dimensionality of a system whose scale exceeds molecular systems so-far considered with similar tools. This bypasses a need for researcher speculation on what elements of a system best contribute to summarizing major transitions and suggests latent models are effective and insightful when applied to large systems with a diversity of complex behaviors.

Chiral structure fluctuations predicted by a coarse-grained model of peptide aggregation

This work reports on the chiral structure fluctuations of peptide clusters at the early stages of aggregation in a coarse-grained peptide model. Our model reproduces a variety of aggregate structures, from disordered to crystal-like, that are observed experimentally. Unexpectedly, our molecular dynamics simulations showed that the small peptide cluster undergoes chiral structure fluctuations although the underlying implicit solvent model does not assume the chirality of peptides. The chiral fluctuations are quantified through a cluster twist parameter. A simple model is presented where the twist parameter undergoes a stochastic diffusion on a 1D potential surface. The shape of the potential surface changes with the cluster size. The model shows semi-quantitative agreement with the simulations. We hypothesize that the chiral fluctuations at the early stages of peptide aggregation can contribute to the selection of the final fibril structures.

Amyloid assembly is dominated by misregistered kinetic traps on an unbiased energy landscape

Atomistic description of protein fibril formation has been elusive due to the complexity and long time scales of the conformational search. Here, we develop a multiscale approach combining numerous atomistic simulations in explicit solvent to construct Markov State Models (MSMs) of fibril growth. The search for the in-register fully bound fibril state is modeled as a random walk on a rugged two-dimensional energy landscape defined by β-sheet alignment and hydrogen-bonding states, whereas transitions involving states without hydrogen bonds are derived from kinetic clustering. The reversible association/dissociation of an incoming peptide and overall growth kinetics are then computed from MSM simulations. This approach is applied to derive a parameter-free, comprehensive description of fibril elongation of Aβ_16-22 and how it is modulated by phenylalanine-to-cyclohexylalanine (CHA) mutations. The trajectories show an aggregation mechanism in which the peptide spends most of its time trapped in misregistered β-sheet states connected by weakly bound states twith short lifetimes. Our results recapitulate the experimental observation that mutants CHA19 and CHA1920 accelerate fibril elongation but have a relatively minor effect on the critical concentration for fibril growth. Importantly, the kinetic consequences of mutations arise from cumulative effects of perturbing the network of productive and nonproductive pathways of fibril growth. This is consistent with the expectation that nonfunctional states will not have evolved efficient folding pathways and, therefore, will require a random search of configuration space. This study highlights the importance of describing the complete energy landscape when studying the elongation mechanism and kinetics of protein fibrils.

Determination of the Aggregate Binding Site of Amyloid Protofibrils Using Electron Capture Dissociation Tandem Mass Spectrometry

Amyloid fibril formation is a hallmark in a range of human diseases. Analysis of the molecular details of amyloid aggregation, however, is limited by the difficulties in solubilizing, separating, and identifying the aggregated biomolecules. Additional labeling or protein modification is required in many current analytical techniques in order to provide molecular details of amyloid protein aggregation, but these modifications may result in protein structure disruption. Herein, ultrahigh resolution mass spectrometry (MS) with electron capture dissociation tandem MS (ECD MS/MS) has been applied to monitor the formation of early oligomers of human islet amyloid polypeptide (hIAPP), which aggregate rapidly in the pancreas of type II diabetes (T2D) patients. ECD MS/MS results show the aggregation region of the early oligomers is at the Ser-28/Ser-29 residue of a hIAPP unit and at the Asn-35 residue of another hIAPP unit near the C-terminus in the gas phase. These data contribute to the understanding of the binding site between hIAPP units which may help for specific target region therapeutic development in the future. Furthermore, MS has also been applied to quantify the amount of soluble amyloid protein remaining in the incubated solutions, which can be used to estimate the aggregation rate of amyloid protein during incubation (28 days). These data are further correlated with the results obtained using fluorescence spectroscopy and transmission electron microscopy (TEM) to generate a general overview of amyloid protein aggregation. The methods demonstrated in this article not only explore the aggregation site of hIAPP down to an amino acid residue level, but are also applicable to many amyloid protein aggregation studies.

Theoretical and computational advances in protein misfolding

Misfolded proteins escape the cellular quality control mechanism and fail to fold properly or remain correctly folded leading to a loss in their functional specificity. Thus misfolding of proteins cause a large number of very different diseases ranging from errors in metabolism to various types of complex neurodegenerative diseases. A theoretical and computational perspective of protein misfolding is presented with a special emphasis on its salient features, mechanism and consequences. These insights quantitatively analyze different determinants of misfolding, that may be applied to design disease specific molecular targets.

Conformational dynamics of amyloid-β (16–22) peptide in aqueous ionic liquids

Molecular dynamics simulations of amyloid-β (16–22) peptide dimer in water as well as at two different experimentally studied concentrations of hydrated ionic liquids (ILs), ethylammonium mesylate (EAM), ethylammonium nitrate (EAN), and triethylammonium mesylate (TEAM), were carried out employing an umbrella sampling method. We used the average Ψ angle of the peptide backbone as the reaction coordinate to observe the conformational changes of a peptide dimer. Secondary structural element values were calculated for the peptide dimer along the reaction coordinate to see the transition of the peptide dimer between β-sheet and α-helix conformations. We observe the β-sheet conformation as the global minimum on the free energy surfaces in both EAM and EAN ILs at both the concentrations and at a low concentration of TEAM. However, we observe α-helix conformation as the global minimum at a high concentration of TEAM. Our results are in good correlation with the experimental findings. We calculated the average number of intramolecular and intermolecular hydrogen bonds of α-helix and β-sheet conformations in all solutions, and they are in correlation with the secondary structure element values. To understand the peptide–IL interactions, atom–atom radial distribution functions of cation, anion, and water around amide oxygen and hydrogen atoms were calculated. The solvent-accessible surface area of the peptide dimer was calculated to understand the exposure of the peptide towards the solvent during conformational changes. Finally, van der Waals (vdW) and Coulomb interaction energies were calculated between peptide–cation, peptide–anion, and peptide–water to understand the stability of conformations in different concentrations. We find that the TEA cation has more vdW interaction energy compared to Coulomb interaction energy with peptide in 70% (w/w) TEAM, which mimics a membrane-like environment to induce α-helix conformation rather than β-sheet conformation.

Molecular dynamics simulations of amyloid-β (16–22) peptide dimer at two different experimentally studied concentrations of hydrated ethylammonium mesylate, ethylammonium nitrate, and triethylammonium mesylate were carried out employing an umbrella sampling method.

Bleomycin modulates amyloid aggregation in β-amyloid and hIAPP

Aberrant misfolding and amyloid aggregation, which result in amyloid fibrils, are frequent and critical pathological incidents in various neurodegenerative disorders. Multiple drugs or inhibitors have been investigated to avert amyloid aggregation in individual peptides, exhibiting sequence-dependent inhibition mechanisms. Establishing or inventing inhibitors capable of preventing amyloid aggregation in a wide variety of amyloid peptides is quite a daunting task. Bleomycin (BLM), a complex glycopeptide, has been widely used as an antibiotic and antitumor drug due to its ability to inhibit DNA metabolism, and as an antineoplastic, especially for solid tumors. In this study, we investigated the dual inhibitory effects of BLM on Aβ aggregation, associated with Alzheimer's disease and hIAPP, which is linked to type 2 diabetes, using both computational and experimental techniques. Combined results from drug repurposing and replica exchange molecular dynamics simulations demonstrate that BLM binds to the β-sheet region considered a hotspot for amyloid fibrils of Aβ and hIAPP. BLM was also found to be involved in β-sheet destabilization and, ultimately, in its reduction. Further, experimental validation through in vitro amyloid aggregation assays was obtained wherein the fibrillar load was decreased for the BLM-treated Aβ and hIAPP peptides in comparison to controls. For the first time, this study shows that BLM is a dual inhibitor of Aβ and hIAPP amyloid aggregation. In the future, the conformational optimization and processing of BLM may help develop various efficient sequence-dependent inhibitors against amyloid aggregation in various amyloid peptides.

Bleomycin acts as a dual inhibitor against both amyloid β and human islet amyloid polypeptide by binding to the β-sheet grooves considered as the amyloids hotspot.

Molecular Model for the Self-Assembly of the Cyclic Lipodepsipeptide Pseudodesmin A

Self-assembly of peptides into supramolecular structures represents an active field of research with potential applications ranging from material science to medicine. Their study typically involves the application of a large toolbox of spectroscopic and imaging techniques. However, quite often, the structural aspects remain underexposed. Besides, molecular modeling of the self-assembly process is usually difficult to handle, since a vast conformational space has to be sampled. Here, we have used an approach that combines short molecular dynamics simulations for peptide dimerization and NMR restraints to build a model of the supramolecular structure from the dimeric units. Experimental NMR data notably provide crucial information about the conformation of the monomeric units, the supramolecular assembly dimensions, and the orientation of the individual peptides within the assembly. This in silico/in vitro mixed approach enables us to define accurate atomistic models of supramolecular structures of the bacterial cyclic lipodepsipeptide pseudodesmin A.

A hybrid, bottom-up, structurally accurate, Go¯-like coarse-grained protein model

Coarse-grained (CG) protein models in the structural biology literature have improved over the years from being simple tools to understand general folding and aggregation driving forces to capturing detailed structures achieved by actual folding sequences. Here, we ask whether such models can be developed systematically from recent advances in bottom-up coarse-graining methods without relying on bioinformatic data (e.g., protein data bank statistics). We use relative entropy coarse-graining to develop a hybrid CG but Go¯-like CG peptide model, hypothesizing that the landscape of proteinlike folds is encoded by the backbone interactions, while the sidechain interactions define which of these structures globally minimizes the free energy in a unique native fold. To construct a model capable of capturing varied secondary structures, we use a new extended ensemble relative entropy method to coarse-grain based on multiple reference atomistic simulations of short polypeptides with varied α and β character. Subsequently, we assess the CG model as a putative protein backbone forcefield by combining it with sidechain interactions based on native contacts but not incorporating native distances explicitly, unlike standard Go¯ models. We test the model's ability to fold a range of proteins and find that it achieves high accuracy (∼2 Å root mean square deviation resolution for both short sequences and large globular proteins), suggesting the strong role that backbone conformational preferences play in defining the fold landscape. This model can be systematically extended to non-natural amino acids and nonprotein polymers and sets the stage for extensions to non-Go¯ models with sequence-specific sidechain interactions.

Multiscale Aggregation of the Amyloid Aβ16-22 Peptide: From Disordered Coagulation and Lateral Branching to Amorphous Prefibrils

In this work we investigate the multiscale dynamics of the aggregation process of an amyloid peptide, Aβ_16-22. By performing massive coarse-grained simulations at the quasi-atomistic resolution and including hydrodynamic effects, we followed the formation and growth of a large elongated aggregate and its slow structuring. The elongation proceeds via a two-step nucleation mechanism with disordered aggregates formed initially and subsequently fusing to elongate the amorphous prefibril. A variety of coagulation events coexist, including lateral growth. The latter mechanism, sustained by long-range hydrodynamics correlations, actually can create a large branched structure spanning a few tens of nanometers. Our findings confirm the experimental hypothesis of a critical contribution of lateral growth to the amyloid aggregation kinetics and the capability of our model to sample critical structures like prefibril hosting annular pores.

Top-down Multiscale Approach To Simulate Peptide Self-Assembly from Monomers

Modeling peptide assembly from monomers on large time and length scales is often intractable at the atomistic resolution. To address this challenge, we present a new approach which integrates coarse-grained (CG), mixed-resolution, and all-atom (AA) modeling in a single simulation. We simulate the initial encounter stage with the CG model, while the further assembly and reorganization stages are simulated with the mixed-resolution and AA models. We have implemented this top-down approach with new tools to automate model transformations and to monitor oligomer formations. Further, a theory was developed to estimate the optimal simulation length for each stage using a model peptide, melittin. The assembly level, the oligomer distribution, and the secondary structures of melittin simulated by the optimal protocol show good agreement with prior experiments and AA simulations. Finally, our approach and theory have been successfully validated with three amyloid peptides (β-amyloid 16-22, GNNQQNY fragment from the yeast prion protein SUP35, and α-synuclein fibril 35-55), which highlight the synergy from modeling at multiple resolutions. This work not only serves as proof of concept for multiresolution simulation studies but also presents practical guidelines for further self-assembly simulations at more physically and chemically relevant scales.

Lipid composition and salt concentration as regulatory factors of the anion selectivity of VDAC studied by coarse-grained molecular dynamics simulations

The voltage-dependent anion channel (VDAC) is a mitochondrial outer membrane protein whose fundamental function is to facilitate and regulate the flow of metabolites between the cytosol and the mitochondrial intermembrane space. Using coarse-grained molecular dynamics simulations, we investigated the dependence of VDAC selectivity towards small inorganic anions on two factors: the ionic strength and the lipid composition. In agreement with experimental data we found that VDAC becomes less anion selective with increasing salt concentration due to the screening of a few basic residues that point into the pore lumen. The molecular dynamics simulations provide insight into the regulation mechanism of VDAC selectivity by the composition in the lipid membrane and suggest that the ion distribution is differently modulated by POPE compared to the POPC bilayer. This occurs through the more persistent interactions of acidic residues located at both rims of the β-barrel with head groups of POPE which in turn impact the electrostatic potential and thereby the selectivity of the pore. This mechanism occurs not only in POPE single component membranes but also in a mixed POPE/POPC bilayer by an enrichment of POPE over POPC lipids on the surface of VDAC. Thus we show here that computationally-inexpensive coarse-grained simulations are able to capture, in a semi-quantitative way, essential features of VDAC anion selectivity and could pave the way toward a molecular level understanding of metabolite transport in natural membranes.

Encoding and selecting coarse-grain mapping operators with hierarchical graphs

Coarse-grained (CG) molecular dynamics (MD) can simulate systems inaccessible to fine-grained (FG) MD simulations. A CG simulation decreases the degrees of freedom by mapping atoms from an FG representation into agglomerate CG particles. The FG to CG mapping is not unique. Research into systematic selection of these mappings is challenging due to their combinatorial growth with respect to the number of atoms in a molecule. Here we present a method of reducing the total count of mappings by imposing molecular topology and symmetry constraints. The count reduction is illustrated by considering all mappings for nearly 50 000 molecules. The resulting number of mapping operators is still large, so we introduce a novel hierarchical graphical approach which encodes multiple CG mapping operators. The encoding method is demonstrated for methanol and a 14-mer peptide. With the test cases, we show how the encoding can be used for automated selection of reasonable CG mapping operators.

Characterisation of protein aggregation with the Smoluchowski coagulation approach for use in biopharmaceuticals

Protein aggregation is a field of increasing importance in the biopharmaceutical industry. Aggregated particles decrease the effectiveness of the drug and are associated with other risks, such as increased immunogenicity. This article explores the possibility of using the Smoluchowski coagulation equation and similar models in the prediction of aggregate-particle formation. Three different monoclonal antibodies, exhibiting different aggregation pathways, are analysed. Experimental data are complemented with aggregation dynamics calculated by a coagulation model. Different processes are implemented in the coagulation equation approach, needed to cover the actual phenomena observed in the aggregation of biopharmaceuticals, such as the initial conformational change of the native monomer and reversibility of smaller oligomers. When describing the formation of larger particles, the effect of different aggregation kernel parameters on the corresponding particle size distribution is studied. A significant impact of the aggregate fractal nature on overall particle size distribution is also analysed. More generally, this work is aimed to establish a mesoscopic phenomenological approach for characterisation of protein aggregation phenomena in the context of biopharmaceuticals, capable of covering various aggregate size scales from nanometres to micrometres and reach large time-scales, up to years, as needed for drug development.

Genistein: A Dual Inhibitor of Both Amyloid β and Human Islet Amylin Peptides

Abnormal misfolding and aggregation of amyloid peptides into amyloid fibrils are common and critical pathological events in many neurodegenerative diseases. Most inhibitors or drugs have been developed to prevent amyloid aggregation of a specific peptide, showing sequence-dependent inhibition mechanisms. It is more challenging to develop or discover inhibitors capable of preventing the aggregation of two or more different amyloid peptides. Genistein, a major phytoestrogen in soybean, has been widely used as an anti-inflammation and cerebrovascular drug due to its antioxidation and antiacetylcholinesterase effects. Herein, we examine the inhibitory effects of genistein on the aggregation of amyloid-β (Aβ, associated with Alzheimer's disease) and human islet amylin (hIAPP, associated with type 2 diabetes) and Aβ- and hIAPP-induced neurotoxicity using a combination of experimental and computational approaches. Collective experimental results from thioflavin T (ThT), atomic force microscopy (AFM), and circular dichroism (CD) demonstrate that genistein shows strong inhibition ability to prevent the conformational transition of both Aβ and hIAPP monomers to β-sheet structures, thus reducing final amyloid fibrillization from Aβ and hIAPP monomer aggregation by 40-63%. Further 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT), lactate dehydrogenase (LDH), and large unilamellar vesicle (LUV) assays show that genistein helps to increase cell viability, decrease cell apoptosis, and reduce cell membrane leakage, where the cell protection effect of genistein is likely correlated with its reduced membrane leakage. Comparative molecular dynamics (MD) simulations reveal that genistein prefers to bind the β-sheet groove, a common structural motif of amyloid fibrils, of both Aβ and hIAPP oligomers to interfere with their self-aggregation. This work for the first time demonstrates genistein as a dual inhibitor of Aβ and hIAPP aggregation. Further structural optimization and refinement of genistein may generate a series of effective sequence-independent inhibitors against the aggregation and toxicity of different amyloid peptides.

Intrinsic Conformational Preferences and Interactions in α-Synuclein Fibrils: Insights from Molecular Dynamics Simulations

Amyloid formation by the intrinsically disordered α-synuclein protein is the hallmark of Parkinson's disease. We present atomistic Molecular Dynamics simulations of the core of α-synuclein using enhanced sampling techniques to describe the conformational and binding free energy landscapes of fragments implicated in fibril stabilization. The theoretical framework is derived to combine the free energy profiles of the fragments into the reaction free energy of a protein binding to a fibril. Our study shows that individual fragments in solution have a propensity toward attaining non-β conformations, indicating that in a fibril β-strands are stabilized by interactions with other strands. We show that most dimers of hydrogen-bonded fragments are unstable in solution, while hydrogen bonding stabilizes the collective binding of five fragments to the end of a fibril. Hydrophobic effects make further contributions to the stability of fibrils. This study is the first of its kind where structural and binding preferences of the five major fragments of the hydrophobic core of α-synuclein have been investigated. This approach improves sampling of intrinsically disordered proteins, provides information on the binding mechanism between the core sequences of α-synuclein, and enables the parametrization of coarse grained models.

Amyloid growth and membrane damage: Current themes and emerging perspectives from theory and experiments on Aβ and hIAPP

Alzheimer's Disease (AD) and Type 2 diabetes mellitus (T2DM) are two incurable diseases both hallmarked by an abnormal deposition of the amyloidogenic peptides Aβ and Islet Amyloid Polypeptide (IAPP) in affected tissues. Epidemiological data demonstrate that patients suffering from diabetes are at high risk of developing AD, thus making the search for factors common to the two pathologies of special interest for the design of new therapies. Accumulating evidence suggests that the toxic properties of both Aβ or IAPP are ascribable to their ability to damage the cell membrane. However, the molecular details describing Aβ or IAPP interaction with membranes are poorly understood. This review focuses on biophysical and in silico studies addressing these topics. Effects of calcium, cholesterol and membrane lipid composition in driving aberrant Aβ or IAPP interaction with the membrane will be specifically considered. The cross correlation of all these factors appears to be a key issue not only to shed light in the countless and often controversial reports relative to this area but also to gain valuable insights into the central events leading to membrane damage caused by amyloidogenic peptides. This article is part of a Special Issue entitled: Protein Aggregation and Misfolding at the Cell Membrane Interface edited by Ayyalusamy Ramamoorthy.

Molecular simulation aspects of amyloid peptides at membrane interface

The interactions of amyloid peptides with cell membranes play an important role in maintaining the integrity and functionality of cell membrane. A thorough molecular-level understanding of the structure, dynamics, and interactions between amyloid peptides and cell membranes is critical to amyloid aggregation and toxicity mechanisms for the bench-to-bedside applications. Here we review the most recent computational studies of amyloid peptides at model cell membranes. Different mechanisms of action of amyloid peptides on/in cell membranes, targeted by different computational techniques at different lengthscales and timescales, are rationally discussed. Finally, we have proposed some new insights into the remaining challenges and perspectives for future studies to improve our understanding of the activity of amyloid peptides associated with protein-misfolding diseases. This article is part of a Special Issue entitled: Protein Aggregation and Misfolding at the Cell Membrane Interface edited by Ayyalusamy Ramamoorthy.