A Perspective on the Prospective Use of AI in Protein Structure Prediction.

AlphaFold2 (AF2) and RoseTTaFold (RF) have revolutionized structural biology, serving as highly reliable and effective methods for predicting protein structures. This article explores their impact and limitations, focusing on their integration into experimental pipelines and their application in diverse protein classes, including membrane proteins, intrinsically disordered proteins (IDPs), and oligomers. In experimental pipelines, AF2 models help X-ray crystallography in resolving the phase problem, while complementarity with mass spectrometry and NMR data enhances structure determination and protein flexibility prediction. Predicting the structure of membrane proteins remains challenging for both AF2 and RF due to difficulties in capturing conformational ensembles and interactions with the membrane. Improvements in incorporating membrane-specific features and predicting the structural effect of mutations are crucial. For intrinsically disordered proteins, AF2's confidence score (pLDDT) serves as a competitive disorder predictor, but integrative approaches including molecular dynamics (MD) simulations or hydrophobic cluster analyses are advocated for accurate dynamics representation. AF2 and RF show promising results for oligomeric models, outperforming traditional docking methods, with AlphaFold-Multimer showing improved performance. However, some caveats remain in particular for membrane proteins. Real-life examples demonstrate AF2's predictive capabilities in unknown protein structures, but models should be evaluated for their agreement with experimental data. Furthermore, AF2 models can be used complementarily with MD simulations. In this Perspective, we propose a "wish list" for improving deep-learning-based protein folding prediction models, including using experimental data as constraints and modifying models with binding partners or post-translational modifications. Additionally, a meta-tool for ranking and suggesting composite models is suggested, driving future advancements in this rapidly evolving field.

Lessons learned from urgent computing in Europe: Tackling the COVID-19 pandemic.

PRACE (Partnership for Advanced Computing in Europe), an international not-for-profit association that brings together the five largest European supercomputing centers and involves 26 European countries, has allocated more than half a billion core hours to computer simulations to fight the COVID-19 pandemic. Alongside experiments, these simulations are a pillar of research to assess the risks of different scenarios and investigate mitigation strategies. While the world deals with the subsequent waves of the pandemic, we present a reflection on the use of urgent supercomputing for global societal challenges and crisis management.

Proton- versus Cation-Selective Transport of Saccharide Rim-Appended Pillar[5]arene Artificial Water Channels.

Transport of water across cell membranes is a fundamental process for important biological functions. Herein, we focused our research on a new type of symmetrical saccharide rim-functionalized pillar[5]arene (PA-S) artificial water channels with variable pore structures. To point out the versatility of PA-S channels, we systematically varied the nature of anchoring/gate keepers d-mannoside, d-mannuronic acid, or sialic acid H-bonding groups on lateral pillar[5]arene (PA) arms, known as good membrane adhesives, to best describe the influence of the chemical structure on their transport activity. The control of hydrophobic membrane binding-hydrophilic water binding balance is an important feature influencing the channels' structuration and efficiency for a proper insertion into bilayer membranes. The glycosylated PA channels' transport performances were assessed in lipid bilayer membranes, and the channels were able to transport water at high rates (∼106-107 waters/s/channel within 1 order of magnitude as for aquaporins), serving as selective proton railways with total Na+ and K+ rejection. Molecular simulation substantiates the idea that the PAs can generate supramolecular pores, featuring hydrophilic carbohydrate gate-keepers that serve as water-sponge relays at the channel entrance, effectively absorbing and redirecting water within the channel. The present channels may be regarded as a rare biomimetic example of artificial channels presenting proton vs cation transport selectivity performances.

Combinatorial Screening of Water/Proton Permeation of Self-Assembled Pillar[5]arene Artificial Water Channel Libraries.

Artificial water channels (AWCs) that selectively transport water and reject ions through bilayer membranes have potential to act as synthetic Aquaporins (AQPs). AWCs can have a similar osmotic permeability, better stability, with simpler manufacture on a larger-scale and have higher functional density and surface permeability when inserted into the membrane. Here, we report the screening of combinatorial libraries of symmetrical and unsymmetrical rim-functionalized PAs A-D that are able to transport ca. 107 -108 water molecules/s/channel, which is within 1 order of magnitude of AQPs' and show total ion and proton rejection. Among the four channels, C and D are 3-4 times more water permeable than A and B when inserted in bilayer membranes. The binary combinations of A-D with different molar ratios could be expressed as an independent (linear ABA), a recessive (inhibition AB, AC, DB, ACA), or a dominant (amplification, DBD) behavior of the water net permeation events.

A large disordered region confers a wide spanning volume to vertebrate Suppressor of Fused as shown in a trans-species solution study.

Hedgehog (Hh) pathway inhibition by the conserved protein Suppressor of Fused (SuFu) is crucial to vertebrate development. By constrast, SuFu loss-of-function mutant has little effect in drosophila. Previous publications showed that the crystal structures of human and drosophila SuFu consist of two ordered domains that are capable of breathing motions upon ligand binding. However, the crystal structure of human SuFu does not give information about twenty N-terminal residues (IDR1) and an eighty-residue-long region predicted as disordered (IDR2) in the C-terminus, whose function is important for the pathway repression. These two intrinsically disordered regions (IDRs) are species-dependent. To obtain information about the IDR regions, we studied full-length SuFu's structure in solution, both with circular dichroism and small angle X-ray scattering, comparing drosophila, zebrafish and human species, to better understand this considerable difference. Our studies show that, in spite of similar crystal structures restricted to ordered domains, drosophila and vertebrate SuFu have very different structures in solution. The IDR2 of vertebrates spans a large area, thus enabling it to reach for partners and be accessible for post-translational modifications. Furthermore, we show that the IDR2 region is highly conserved within phyla but varies in length and sequence, with insects having a shorter disordered region while that of vertebrates is broad and mobile. This major variation may explain the different phenotypes observed upon SuFu removal.

Between Two Walls: Modeling the Adsorption Behavior of ß-Glucosidase A on Bare and SAM-Functionalized Gold Surfaces.

The efficient immobilization of enzymes on surfaces remains a complex but central issue in the biomaterials field, which requires us to understand this process at the atomic level. Using a multiscale approach combining all-atom molecular dynamics and coarse-grain Brownian dynamics simulations, we investigated the adsorption behavior of ß-glucosidase A (ßGA) on bare and self-assembled monolayer (SAM)-functionalized gold surfaces. We monitored the enzyme position and orientation during the molecular dynamics (MD) trajectories and measured the contacts it forms with both surfaces. While the adsorption process has little impact on the protein conformation, it can nonetheless perturb its mechanical properties and catalytic activity. Our results show that compared to the SAM-functionalized surface, the adsorption of ßGA on bare gold is more stable, but less specific, and more likely to disrupt the enzyme's function. This observation emphasizes the fact that the structural organization of proteins at the solid interface is a key point when designing devices based on enzyme immobilization, as one must find an acceptable stability-activity trade-off.

Artificial Water Channels Form Precursors to Sponge-Like Aggregates in Water-Ethanol Mixtures.

Self-assembled artificial water channels (AWCs) are reshaping current water desalination technologies. Recently, the improvements achieved by incorporating hydrophilic compounds into polyamide membranes (PA) at the interface were confirmed experimentally. However, the determination of the nanoscale structures of AWCs remains unclear. An important step in the preparation of PA membranes is the solubilization of a colloidal suspension of the solid phase in a water-ethanol mixture. We perform molecular dynamics simulations to study the nanoscale structures of AWC aggregates. We characterize the size and shape of the aggregates at several key locations in the ternary phase diagram. The role of ethanol in the formation of the interface between the solvent and the solute phase is highlighted. We found that the structure of the aggregates formed in the ternary solution resembled the disordered sponge-like structures observed when AWCs were inserted into lipid membranes. Such permeable sponge architectures allow the passage of water despite their noncrystalline organization and were previously shown to be consistent with AWC permeation measurements in membrane environments.

Glutathionylation primes soluble glyceraldehyde-3-phosphate dehydrogenase for late collapse into insoluble aggregates.

Protein aggregation is a complex physiological process, primarily determined by stress-related factors revealing the hidden aggregation propensity of proteins that otherwise are fully soluble. Here we report a mechanism by which glycolytic glyceraldehyde-3-phosphate dehydrogenase of Arabidopsis thaliana (AtGAPC1) is primed to form insoluble aggregates by the glutathionylation of its catalytic cysteine (Cys149). Following a lag phase, glutathionylated AtGAPC1 initiates a self-aggregation process resulting in the formation of branched chains of globular particles made of partially misfolded and totally inactive proteins. GSH molecules within AtGAPC1 active sites are suggested to provide the initial destabilizing signal. The following removal of glutathione by the formation of an intramolecular disulfide bond between Cys149 and Cys153 reinforces the aggregation process. Physiological reductases, thioredoxins and glutaredoxins, could not dissolve AtGAPC1 aggregates but could efficiently contrast their growth. Besides acting as a protective mechanism against overoxidation, S-glutathionylation of AtGAPC1 triggers an unexpected aggregation pathway with completely different and still unexplored physiological implications.

Hydroxy Channels-Adaptive Pathways for Selective Water Cluster Permeation.

Artificial water channels (AWCs) are known to selectively transport water, with ion exclusion. Similarly to natural porins, AWCs encapsulate water wires or clusters, offering continuous and iterative H-bonding that plays a vital role in their stabilization. Herein, we report octyl-ureido-polyol AWCs capable of self-assembly into hydrophilic hydroxy channels. Variants of ethanol, propanediol, and trimethanol are used as head groups to modulate the water transport permeabilities, with rejection of ions. The hydroxy channels achieve a single-channel permeability of 2.33 × 108 water molecules per second, which is within the same order of magnitude as the transport rates for aquaporins. Depending on their concentration in the membrane, adaptive channels are observed in the membrane. Over increased concentrations, a significant shift occurs, initiating unexpected higher water permeation. Molecular simulations probe that spongelike or cylindrical aggregates can form to generate transient cluster water pathways through the bilayer. Altogether, the adaptive self-assembly is a key feature influencing channel efficiency. The adaptive channels described here may be considered an important milestone contributing to the systematic discovery of artificial water channels for water desalination.

Molecular dynamics simulations reveal statistics and microscopic mechanisms of water permeation in membrane-embedded artificial water channel nanoconstructs.

Understanding water transport mechanisms at the nanoscale level remains a challenge for theoretical chemical physics. Major advances in chemical synthesis have allowed us to discover new artificial water channels, rivaling with or even surpassing water conductance and selectivity of natural protein channels. In order to interpret experimental features and understand microscopic determinants for performance improvements, numerical approaches based on all-atom molecular dynamics simulations and enhanced sampling methods have been proposed. In this study, we quantify the influence of microscopic observables, such as channel radius and hydrogen bond connectivity, and of meso-scale features, such as the size of self-assembly blocks, on the permeation rate of a self-assembled nanocrystal-like artificial water channel. Although the absolute permeation rate extrapolated from these simulations is overestimated by one order of magnitude compared to the experimental measurement, the detailed analysis of several observed conductive patterns in large assemblies opens new pathways to scalable membranes with enhanced water conductance for the future design.

Visualizing protein structures - tools and trends.

Molecular visualization is fundamental in the current scientific literature, textbooks and dissemination materials. It provides an essential support for presenting results, reasoning on and formulating hypotheses related to molecular structure. Tools for visual exploration of structural data have become easily accessible on a broad variety of platforms thanks to advanced software tools that render a great service to the scientific community. These tools are often developed across disciplines bridging computer science, biology and chemistry. This mini-review was written as a short and compact overview for scientists who need to visualize protein structures and want to make an informed decision which tool they should use. Here, we first describe a few 'Swiss Army knives' geared towards protein visualization for everyday use with an existing large user base, then focus on more specialized tools for peculiar needs that are not yet as broadly known. Our selection is by no means exhaustive, but reflects a diverse snapshot of scenarios that we consider informative for the reader. We end with an account of future trends and perspectives.

Characterization of ß-turns by electronic circular dichroism spectroscopy: a coupled molecular dynamics and time-dependent density functional theory computational study.

Electronic circular dichroism is one of the most used spectroscopic techniques for peptide and protein structural characterization. However, while valuable experimental spectra exist for α-helix, ß-sheet and random coil secondary structures, previous studies showed important discrepancies for ß-turns, limiting their use as a reference for structural studies. In this paper, we simulated circular dichroism spectra for the best-characterized ß-turns in peptides, namely types I, II, I' and II'. In particular, by combining classical molecular dynamics simulations and state-of-the-art quantum time-dependent density functional theory (with the polarizable embedding multiscale model) computations, two common electronic circular dichroism patterns were found for couples of ß-turn types (namely, type I/type II' and type II/type I'), at first for a minimal di-peptide model (Ace-Ala-Ala-NHMe), but also for all sequences tested with non-aromatic residues in the central positions. On the other hand, as expected, aromatic substitution causes important perturbations to the previously found ECD patterns. Finally, by applying suitable approximations, these patterns were subsequently rationalized based on the exciton chirality rule. All these results provide useful predictions and pave the way for a possible experimental characterization of ß-turns based on circular dichroism spectroscopy.

String method solution of the gating pathways for a pentameric ligand-gated ion channel.

Pentameric ligand-gated ion channels control synaptic neurotransmission by converting chemical signals into electrical signals. Agonist binding leads to rapid signal transduction via an allosteric mechanism, where global protein conformational changes open a pore across the nerve cell membrane. We use all-atom molecular dynamics with a swarm-based string method to solve for the minimum free-energy gating pathways of the proton-activated bacterial GLIC channel. We describe stable wetted/open and dewetted/closed states, and uncover conformational changes in the agonist-binding extracellular domain, ion-conducting transmembrane domain, and gating interface that control communication between these domains. Transition analysis is used to compute free-energy surfaces that suggest allosteric pathways; stabilization with pH; and intermediates, including states that facilitate channel closing in the presence of an agonist. We describe a switching mechanism that senses proton binding by marked reorganization of subunit interface, altering the packing of ß-sheets to induce changes that lead to asynchronous pore-lining M2 helix movements. These results provide molecular details of GLIC gating and insight into the allosteric mechanisms for the superfamily of pentameric ligand-gated channels.

Biomimetic Approach for Highly Selective Artificial Water Channels Based on Tubular Pillar[5]arene Dimers.

Artificial water channels mimicking natural aquaporins (AQPs) can be used for selective and fast transport of water. Here, we quantify the transport performances of peralkyl-carboxylate-pillar[5]arenes dimers in bilayer membranes. They can transport ≈107 water molecules/channel/second, within one order of magnitude of the transport rates of AQPs, rejecting Na+ and K+ cations. The dimers have a tubular structure, superposing pillar[5]arene pores of 5 Šdiameter with twisted carboxy-phenyl pores of 2.8 Šdiameter. This biomimetic platform, with variable pore dimensions within the same structure, offers size restriction reminiscent of natural proteins. It allows water molecules to selectively transit and prevents bigger hydrated cations from passing through the 2.8 Špore. Molecular simulations prove that dimeric or multimeric honeycomb aggregates are stable in the membrane and form water pathways through the bilayer. Over time, a significant shift of the upper vs. lower layer occurs initiating new unexpected water permeation events through toroidal pores.

Dystrophin's central domain forms a complex filament that becomes disorganized by in-frame deletions.

Dystrophin, encoded by the DMD gene, is critical for maintaining plasma membrane integrity during muscle contraction events. Mutations in the DMD gene disrupting the reading frame prevent dystrophin production and result in severe Duchenne muscular dystrophy (DMD); in-frame internal deletions allow production of partly functional internally deleted dystrophin and result in less severe Becker muscular dystrophy (BMD). Many known BMD deletions occur in dystrophin's central domain, generally considered to be a monotonous rod-shaped domain based on the knowledge of spectrin family proteins. However, the effects caused by these deletions, ranging from asymptomatic to severe BMD, argue against the central domain serving only as a featureless scaffold. We undertook structural studies combining small-angle X-ray scattering and molecular modeling in an effort to uncover the structure of the central domain, as dystrophin has been refractory to characterization. We show that this domain appears to be a tortuous and complex filament that is profoundly disorganized by the most severe BMD deletion (loss of exons 45-47). Despite the preservation of large parts of the binding site for neuronal nitric oxide synthase (nNOS) in this deletion, computational approaches failed to recreate the association of dystrophin with nNOS. This observation is in agreement with a strong decrease of nNOS immunolocalization in muscle biopsies, a parameter related to the severity of BMD phenotypes. The structural description of the whole dystrophin central domain we present here is a first necessary step to improve the design of microdystrophin constructs toward the goal of a successful gene therapy for DMD.

A Molecular Perspective on Mitochondrial Membrane Fusion: From the Key Players to Oligomerization and Tethering of Mitofusin.

Mitochondria are dynamic organelles characterized by an ultrastructural organization which is essential in maintaining their quality control and ensuring functional efficiency. The complex mitochondrial network is the result of the two ongoing forces of fusion and fission of inner and outer membranes. Understanding the functional details of mitochondrial dynamics is physiologically relevant as perturbations of this delicate equilibrium have critical consequences and involved in several neurological disorders. Molecular actors involved in this process are large GTPases from the dynamin-related protein family. They catalyze nucleotide-dependent membrane remodeling and are widely conserved from bacteria to higher eukaryotes. Although structural characterization of different family members has contributed in understanding molecular mechanisms of mitochondrial dynamics in more detail, the complete structure of some members as well as the precise assembly of functional oligomers remains largely unknown. As increasing structural data become available, the domain modularity across the dynamin superfamily emerged as a foundation for transfering the knowledge towards less characterized members. In this review, we will first provide an overview of the main actors involved in mitochondrial dynamics. We then discuss recent example of computational methodologies for the study of mitofusin oligomers, and present how the usage of integrative modeling in conjunction with biochemical data can be an asset in progressing the still challenging field of membrane dynamics.

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.

Water permeation across artificial I-quartet membrane channels: from structure to disorder.

Artificial water channels (AWCs) have been designed for water transport across membranes with the aim to mimic the high water permeability observed for biological systems such as aquaporins (∼108-109 water molecules per s per channel), as well as their selectivity to reject ion permeation at the same time. Recent works on designed self-assembling alkylureido-ethylimidazole compounds forming imidazole-quartet channels (I-quartets), have shown both high water permeability and total ionic-rejection. I-quartets are thus promising candidates for further development of AWCs. However, the molecular mechanism of water permeation as well as I-quartet organization and stability in a membrane environment need to be fully understood to guide their optimal design. Here, we use a wide range of all-atom molecular dynamics (MD) simulations and their analysis to understand the structure/activity relationships of the I-quartet channels. Four different types with varying alkyl chain length or chirality have been studied in a complex fully hydrated lipid bilayer environment at both microsecond and nanosecond scale. Microsecond simulations show two distinct behaviors; (i) two out of four systems maintain chiral dipolar oriented water wires, but also undergo a strong reorganization of the crystal shape, (ii) the two other I-quartet channels completely lose the initial organization, nonetheless keeping a water transport activity. Short MD simulations with higher time resolution were conducted to characterize the dynamic properties of water molecules in these model channels and provided a detailed hypothesis on the molecular mechanism of water permeation. The ordered confined water was characterized with quantitative measures of hydrogen-bond life-time and single particle dynamics, showing variability among I-quartet channels. We will further discuss the underlying assumptions, currently based on self-aggregation simulations and crystal patches embedded in lipid bilayer simulations and attempt to describe possible alternative approaches to computationally capture the water permeation mechanism and the self-assembly process of these AWCs.

Allosteric and hyperekplexic mutant phenotypes investigated on an α1 glycine receptor transmembrane structure.

The glycine receptor (GlyR) is a pentameric ligand-gated ion channel (pLGIC) mediating inhibitory transmission in the nervous system. Its transmembrane domain (TMD) is the target of allosteric modulators such as general anesthetics and ethanol and is a major locus for hyperekplexic congenital mutations altering the allosteric transitions of activation or desensitization. We previously showed that the TMD of the human α1GlyR could be fused to the extracellular domain of GLIC, a bacterial pLGIC, to form a functional chimera called Lily. Here, we overexpress Lily in Schneider 2 insect cells and solve its structure by X-ray crystallography at 3.5 Å resolution. The TMD of the α1GlyR adopts a closed-channel conformation involving a single ring of hydrophobic residues at the center of the pore. Electrophysiological recordings show that the phenotypes of key allosteric mutations of the α1GlyR, scattered all along the pore, are qualitatively preserved in this chimera, including those that confer decreased sensitivity to agonists, constitutive activity, decreased activation kinetics, or increased desensitization kinetics. Combined structural and functional data indicate a pore-opening mechanism for the α1GlyR, suggesting a structural explanation for the effect of some key hyperekplexic allosteric mutations. The first X-ray structure of the TMD of the α1GlyR solved here using GLIC as a scaffold paves the way for mechanistic investigation and design of allosteric modulators of a human receptor.

What Can Human-Guided Simulations Bring to RNA Folding?

Inspired by the recent success of scientific-discovery games for predicting protein tertiary and RNA secondary structures, we have developed an open software for coarse-grained RNA folding simulations, guided by human intuition. To determine the extent to which interactive simulations can accurately predict 3D RNA structures of increasing complexity and lengths (four RNAs with 22-47 nucleotides), an interactive experiment was conducted with 141 participants who had very little knowledge of nucleic acids systems and computer simulations, and had received only a brief description of the important forces stabilizing RNA structures. Their structures and full trajectories have been analyzed statistically and compared to standard replica exchange molecular dynamics simulations. Our analyses show that participants gain easily chemical intelligence to fold simple and nontrivial topologies, with little computer time, and this result opens the door for the use of human-guided simulations to RNA folding. Our experiment shows that interactive simulations have better chances of success when the user widely explores the conformational space. Interestingly, providing on-the-fly feedback of the root mean square deviation with respect to the experimental structure did not improve the quality of the proposed models.