Auswahl der wissenschaftlichen Literatur zum Thema „Protein-RNA docking“

AbstractMotivationGiven the importance of protein–ribonucleic acid (RNA) interactions in many biological processes, a variety of docking algorithms have been developed to predict the complex structure from individual protein and RNA partners in the past decade. However, due to the impact of molecular flexibility, the performance of current methods has hit a bottleneck in realistic unbound docking. Pushing the limit, we have proposed a protein-ensemble–RNA docking strategy to explicitly consider the protein flexibility in protein–RNA docking through an ensemble of multiple protein structures, which is referred to as MPRDock. Instead of taking conformations from MD simulations or experimental structures, we obtained the multiple structures of a protein by building models from its homologous templates in the Protein Data Bank (PDB).ResultsOur approach can not only avoid the reliability issue of structures from MD simulations but also circumvent the limited number of experimental structures for a target protein in the PDB. Tested on 68 unbound–bound and 18 unbound–unbound protein–RNA complexes, our MPRDock/DITScorePR considerably improved the docking performance and achieved a significantly higher success rate than single-protein rigid docking whether pseudo-unbound templates are included or not. Similar improvements were also observed when combining our ensemble docking strategy with other scoring functions. The present homology model-based ensemble docking approach will have a general application in molecular docking for other interactions.Availability and implementationhttp://huanglab.phys.hust.edu.cn/mprdock/Supplementary informationSupplementary data are available at Bioinformatics online.

RNA–protein interactions are crucial for such key biological processes as regulation of transcription, splicing, translation, and gene silencing, among many others. Knowing where an RNA molecule interacts with a target protein and/or engineering an RNA molecule to specifically bind to a protein could allow for rational interference with these cellular processes and the design of novel therapies. Here we present a robust RNA–protein fragment pair-based method, termed RnaX, to predict RNA-binding sites. This methodology, which is integrated into the ModelX tool suite (http://modelx.crg.es), takes advantage of the structural information present in all released RNA–protein complexes. This information is used to create an exhaustive database for docking and a statistical forcefield for fast discrimination of true backbone-compatible interactions. RnaX, together with the protein design forcefield FoldX, enables us to predict RNA–protein interfaces and, when sufficient crystallographic information is available, to reengineer the interface at the sequence-specificity level by mimicking those conformational changes that occur on protein and RNA mutagenesis. These results, obtained at just a fraction of the computational cost of methods that simulate conformational dynamics, open up perspectives for the engineering of RNA–protein interfaces.

AbstractMotivationThe main function of protein–RNA interaction is to regulate the expression of genes. Therefore, studying protein–RNA interactions is of great significance. The information of three-dimensional (3D) structures reveals that atomic interactions are particularly important. The calculation method for modeling a 3D structure of a complex mainly includes two strategies: free docking and template-based docking. These two methods are complementary in protein–protein docking. Therefore, integrating these two methods may improve the prediction accuracy.ResultsIn this article, we compare the difference between the free docking and the template-based algorithm. Then we show the complementarity of these two methods. Based on the analysis of the calculation results, the transition point is confirmed and used to integrate two docking algorithms to develop P3DOCK. P3DOCK holds the advantages of both algorithms. The results of the three docking benchmarks show that P3DOCK is better than those two non-hybrid docking algorithms. The success rate of P3DOCK is also higher (3–20%) than state-of-the-art hybrid and non-hybrid methods. Finally, the hierarchical clustering algorithm is utilized to cluster the P3DOCK’s decoys. The clustering algorithm improves the success rate of P3DOCK. For ease of use, we provide a P3DOCK webserver, which can be accessed at www.rnabinding.com/P3DOCK/P3DOCK.html. An integrated protein–RNA docking benchmark can be downloaded from http://rnabinding.com/P3DOCK/benchmark.html.Availability and implementationwww.rnabinding.com/P3DOCK/P3DOCK.html.Supplementary informationSupplementary data are available at Bioinformatics online.

RNA-ligand docking is a part of computational biology, which is currently lowly recognized compared to the protein-ligand docking procedure commonly applied for drug discovery. This in silico study aims to create a simplified protocol for RNA-ligand docking, which is applicable to RNA-targeting small molecular drug screening. Four alkaloids (berberine, colchicine, nicotine, and tomatine) were subjected to this study and contended against the SARS-CoV-2 genomic RNA -1 PRF component targeting control drug, merafloxacin, including two known intercalator berberine and colchicine, a small alkaloid nicotine and a large alkaloid tomatine. The alkaloids were screened for drug-likeness properties (Lipinski’s Rules of 5 or LRo5), bioavailability indexes, and synthetic accessibility values using SwissADME before docking. The docking used PyRx – Autodock Vina and re-scored for RNA-ligand scoring using AnnapuRNA. The docking results have the interactions mapped using fingeRNAt and visualized using Discovery Studio. Molecular dynamics using CHARMM36 and AMBER forcefields were simulated in NAMD. The molecular dynamics 1 ns simulation results showed that the ligand interaction over time did not cause much interference with the RNA, indicated by the low number of RMSD changes between RNA itself and the RNA-ligand complex. Additionally, CHARMM36 forcefield provided more stable fluctuation compared to AMBER. The results indicated that tomatine disobeyed LRo5 and had a low bioavailability index and bad synthetic accessibility value, while the rest alkaloids passed. In the end, berberine has an even higher docking score than the control drug. The study also shows that this protocol can be useful for future RNA-ligand computational studies.

Les complexes protéine-ARN jouent un rôle crucial dans la régulation cellulaire. La prédiction de leur structure 3D a des applications dans la conception de protéines et de médicaments. Le projet ITN RNAct visait à combiner des méthodes expérimentales et informatiques pour concevoir de nouveaux "motifs de reconnaissance de l'ARN" (RRM) - domaines protéiques interagissant avec l'ARN simple brin (ARNsb) - pour la biologie synthétique et la bioanalyse. La modélisation des complexes protéine-ARNsb (amarrage) est ardue car l'ARNsb n'a pas de structure propre dans sa forme libre. L'amarrage traditionnelle échantillonne les positions relatives (poses) de 2 structures moléculaires et les note pour sélectionner les plus probables. Il n'est pas directement applicable ici en raison de l'absence de structures libres d'ARNsb, pas plus que l'apprentissage profond en raison du nombre trop faible de structures connues. L'amarrage par fragments, état de l'art pour l'ARNsb, amarre toutes les conformations possibles de fragments d'ARN sur une protéine et assemble les poses les mieux notées de manière combinatoire. Notre méthode ssRNA'TTRACT utilise le logiciel d'amarrage ATTRACT et sa représentation gros grain qui remplace des groupes d'atomes par une bille. Cependant, les paramètres ARN-protéine de sa fonction de notation (ASF) ne sont pas spécifiques à l'ARNsb et peuvent être optimisés. De plus, des caractéristiques spécifiques aux RRM peuvent être apprises et guider l'amarrage. Nous avons développé un pipeline d'amarrage RRM-ssRNA basé sur les données, pour actualiser une stratégie existante. Les RRM ont 2 acides aminés aromatiques de position conservée, chacun liant par empilement un nucléotide de l'ARN. Mon collègue H. Dhondge a regroupées les structures RRM-ARNsb connues sur critère géométrique et obtenu un ensemble de prototypes de coordonnées 3D de tels empilements dans les RRM. J'ai créé un pipeline qui prend en entrée une séquence de RRM et d'ARN et l'identification des nucléotides empilés, récupère la structure du RRM dans AlphaFoldDB, identifie les positions 3D possibles des nucléotides empilés et exécute ssRNA'TTRACT avec des contraintes de distance maximales vers chaque position. En parallèle, nous avons dérivé HIPPO (HIstogram-based Pseudo-POtential), un potentiel de notation pour les poses gros-grain RRM-ARNsb basé sur la fréquence des distances bille-bille dans les poses quasi-natives versus erronées. HIPPO combine 4 ensembles de paramètres en une note consensus, afin de prendre en compte les divers modes de liaison RRM-ARNsb. Testé dans une approche "leave-one-out", il atteint un enrichissement d'un facteur 3 en quasi-natives dans les 20% de poses mieux notées pour ½ des cas contre ¼ avec ASF, et 'un facteur 4 pour ⅓ des cas contre 7% avec ASF. Surprenamment, HIPPO obtient aussi de meilleurs résultats qu'ASF sur un ensemble test de protéines sans RRM, bien que entraîné sur des RRM. Les approches par fragment rencontrent un problème intrinsèque de notation car certains fragments se lient plus spécifiquement/fortement que d'autres. Or nous avons constaté que, pour le fragment le mieux noté par complexe, HIPPO sélectionne systématiquement plus de quasi-natifs qu'ASF. Cela nous a inspiré une approche d'amarrage incrémentale: chacune des poses bien notées d'un fragment sont utilisées comme graine pour construire une chaîne d'ARN complète de manière incrémentale. Cette stratégie élimine le besoin de contacts conservés connus, jusqu'alors nécessaires pour obtenir des modèles précis, ce qui la rend généralisable aux protéines sans RRM. Nos recherches futures visent à identifier le Η le plus performant pour chaque fragment, potentiellement par apprentissage automatique (profond). Notre approche pour dériver des paramètres de notation est en principe applicable à tout type de protéine/ligand et nous prévoyons de l'étendre à d'autres domaines de protéines liant l'ARN, ainsi qu'à l'ADNsb et aux peptides longs
Protein-RNA complexes play crucial roles in cell regulation. Predicting their 3D structure has applications in protein design and drug development. The ITN project RNAct aimed to combine experimental and computational methods to design new "RNA recognition motifs" (RRM) - protein domains interacting with single-stranded RNA (ssRNA) - for applications in synthetic biology and bioanalysis. Modelling protein-ssRNA complexes (docking) is an arduous task due to the flexibility of ssRNA, which lacks a proper structure in its free form. Traditional docking methods sample the relative positions (poses) of 2 molecular structures and score them to select the correct (near-native) ones. It is not directly applicable here due to the absence of free ssRNA structures, nor is deep learning due to the too low number of known structures for training. Fragment-based docking (FBD), the state-of-the-art approach for ssRNA, docks all possible conformations of RNA fragments onto a protein and assembles their best-scored poses combinatorially. ssRNA'TTRACT, our FBD method, uses the well-known ATTRACT docking software, with its coarse-grained representation that replaces atom groups by one bead. Yet the RNA-protein parameters of ATTRACT scoring function (ASF) are not ssRNA-specific and require optimisation. Additionally, RRM-specific features can be learned and used to guide the docking. With my colleague H. Dhondge, we have developed a data-driven FBD pipeline for RRM-ssRNA complexes, as an updated version of an existing strategy. RRMs have two aromatic amino acids (aa) in conserved positions, each stacking with a nucleotide of the bound ssRNA. H. Dhondge collected all known RRM-ssRNA structures with such stacking and clustered them to obtain a set of prototypes for the 3D coordinates of such interactions in RRM. I then set up a docking pipeline with as input the RRM and RNA sequences and the identification of the stacked nucleotides. The pipeline retrieves the RRM structure from AlphaFoldDB, identifies possible 3D positions of the stacked nucleotides and runs ssRNA'TTRACT with maximal distance restraints toward each position. In parallel, we addressed the weakness of ASF for ssRNA by deriving HIPPO (HIstogram-based Pseudo-POtential), a new scoring potential for ATTRACT poses of ssRNA on RRM, based on the frequency of bead-bead distances in near-native versus wrong poses. It combines 4 distinct parameter sets (four Η) into a consensus scoring, to better account for the diverse RRM-ssRNA binding modes. Tested in a leave-one-out approach, HIPPO reaches a 3-fold enrichment of near-natives in 20% top-scored poses for ½ of the ssRNA fragments, versus ¼ with ASF. It even reaches a 4-fold enrichment for ⅓ of the fragments, versus 7% of the fragments with ASF. Surprisingly, HIPPO performed better than ASF also on a benchmark of non-RRM proteins, while trained only on RRMs. Most FBD approaches encounter inherent scoring issues, probably due to some fragments binding more specifically/strongly than others. To address this point, we examined the best-scored fragment per complex and found that HIPPO consistently selects more near-natives than ASF for this fragment. This inspired an incremental docking approach: the top-ranked poses of one fragment are used as a starting point to build a full RNA chain incrementally. This strategy eliminates the need for known conserved contacts, which have been required so far to obtain accurate models, making it generalizable to non-RRM proteins. Future research aims to identify the best-performing Η for each fragment, potentially using (deep) machine learning. Our workflow to derive scoring parameters is in principle applicable to any protein/ligand type and we plan to expand it to other RNA-binding protein domains, as well as ssDNA and long peptides

Kaposi’s sarcoma-associated herpesvirus, is a double-stranded DNA γ - herpesvirus and the main causative agent of Kaposi’s sarcoma (KS). γ - herpesviruses undergo both lytic and latent replication cycles; and encode proteins that modulate host transcription at the RNA level, by inducing decay of certain mRNAs. Here we describe a mechanism that allows the viral endo-/exonuclease SOX to recognise mRNA targets on the basis of an RNA motif and fold. To induce rapid RNA degradation by subverting the main host mRNA degradation pathway SOX was shown to directly bind Xrn1. This may shed light as to how some viruses evade the host antiviral response and how mRNA degradation processes in the eukaryotic cell are involves in this.

Les interactions ARN-protéine interviennent dans de nombreux processus cellulaires fondamentaux. L'obtention de détails à l'échelle atomique de ces interactions nous éclaire sur leurs fonctions, mais permet également d'envisager la conception rationnelle de ligands pouvant les moduler. Lorsque les deux techniques majeures que sont la RMN et la cristallographie aux rayons X ne permettent pas d'obtenir une structure 3D entre les deux partenaires, des approches de docking peuvent être utilisées pour apporter des modèles. L'application de ces approches aux complexes ARN-protéine se heurtent cependant à une difficulté. Ces complexes résultent en effet souvent de la liaison spécifique d'une courte séquence d'ARN simple-brin (ARNsb) à sa protéine cible. Hors, la flexibilité inhérente aux segments simples-brins impose dans une approche classique de docking d'explorer un large ensemble de leur espace conformationnel. L'objectif du projet est de contourner cette difficulté par le développement d'une approche de docking dite "par fragments". Ce dernier s'est fait à partir de domaines de liaison à l'ARN très représentés dans le monde du vivant. Les résultats ont montré une excellente capacité prédictive de l'approche à partir de la séquence de l'ARN. Ils ont de plus montré un potentiel intéressant dans la prédiction de séquences d'ARN simple-brin préférentiellement reconnues par des domaines de liaisons à l'ARN
RNA-protein interactions mediate numerous fundamental cellular processes. Atomic scale details of these interactions shed light on their functions but can also allow the rational design of ligands that could modulate them. NMR and X-ray crystallography are the 2 main techniques used to resolve 3D highresolution structures between two interacting molecules. Docking approaches can also be utilized to give models as an alternative. However, the application of these approaches to RNA-protein complexes is hampered by an issue. RNA-protein interactions often relies on the specific recognition of a short singlestranded RNA (ssRNA) sequence by the protein. The inherent flexibility of the ssRNA segment would impose, in a classical docking approach, to explore their resulting large conformation space which is not computationally reliable. The goal of this project is to overcome this barrier by using a fragment-based docking approach. This approach developed from some of the most represented RNA-binding domains showed excellent results in the prediction of the ssRNA-protein binding mode from the RNA sequence and also a great potential to predict preferential RNA binding sequences

Macromolecular interactions play very important roles in regulation of all levels of biological processes. Aberrant macromolecular interactions often result in diseases. By applying a combination of spectroscopy, calorimetry, computation and other techniques, the protein-protein interactions in the system of the Shaw2 Kv channel and the protein-RNA interactions in West Nile virus RNA-cellular protein TIAR complex were explored. In the former system, the results shed light on the local structures of the key channel components and their potential interaction mediated by butanol, a general anesthetic. In the later studies, the binding modes of TIAR RRM2 to oligoU RNAs and West Nile virus RNAs were investigated. These findings provided insights into the basis of the specific cellular protein–viral RNA interaction and preliminary data for the development of strategies on how to interfere with virus replication

Molecular docking is a widely used and effective structure-based computational strategy for predicting dynamics between ligands and receptors. Until now the docking software were developed for the protein-ligand interactions and very few docking tools were developed exclusively for the docking of small molecules on the nucleic acid structures like the DNA and RNA. The progress in algorithms and the need for deeper understanding of ligand-nucleic acid interactions more focused, and specialized tools are being developed to explore this hindered area of drug discovery. This chapter is focused on and discus in details about various tools available for docking with nucleic acids and how the rejuvenation of machine learning methods is making its impact on the development of these docking programs.

RNA dependent RNA polymerase (RdRp), an important class of nucleic acid polymerases, encoded by RNA viruses such as SARS-CoV-2, has been a major drug target against viral diseases. Among the twenty-nine SARS-CoV-2 encoded proteins, RdRp is referred to as Non-Structural Protein 12 (Nsp12). Obtaining novel RdRp inhibitors is one of the crucial strategies in developing fast therapeutics against COVID-19. The NCI natural compound database containing 250 thousand small molecules was docked against the available 3D structure of SARS-CoV-2 RdRp (PDB ID: 7BV2). The molecules with best docking scores were assessed for their safety through ADMET predictions. Lead molecules that passed all the parameters of ADMET predictions, were subjected to molecular dynamic simulations and binding affinity analysis. This in silico study gave seven antiviral lead compounds that were having better binding affinities than the remdesivir. Their binding affinities ranged from −8.5 to −11.4 kcal/mol which is significantly higher than remdesivir (−7.59 kcal/mol). Besides, these novel antiviral compounds were found to bind between the Finger and Palm domains, restricting or obstructing the template RNA entry and movement, unlike remdesivir which binds at the 3’ end of the nascent RNA near the palm domain. The data indicates stable structures, favorable binding free energies when compared to remdesivir. Besides having better binding affinity, these leads have easy access to the binding pocket unlike the binding site of remdesivir that is in a deep cleft.These advantages of the identified novel compounds demand for immediate in vitro studies.

This chapter describes viruses as small organisms that retain infectivity after passage through filters small enough to hold back bacteria. Bacteria are measured in terms of the micrometre (μm), which is 10-6 of a metre. For viruses, the nanometre (nm) is used as the unit, which is a thousand times smaller. The chapter points out that viruses are totally dependent on living cells, either eukaryotic or prokaryotic, for their replication and existence, and possess and carry enzymes of their own. Viruses cannot reproduce and amplify and translate into proteins the information in their genomes without the assistance of the cellular architecture and protein translation machinery, namely ribosomes. The chapter discusses viruses that possess only one species of nucleic acid, either DNA or RNA, and have a component for attaching or ‘docking’ to cells so that they can commandeer the cells as virus production factories.

Protein-protein interactions (PPI) are critical for cellular functions, host-pathogen dynamics and are crucial with drug design efforts. The interaction of proteins is dependent on the amino acid sequence of a protein as it determines its binding affinity to various molecules, including drugs, DNA, RNA, and proteins. Polymorphisms, natural DNA variations, affect PPIs by altering protein structure and stability. Computational chemistry is vital for the prediction of ligand-protein interactions through techniques such as docking and molecular dynamics and can elucidate the changes in energy associated with such mutations. We present a user-friendly protocol that uses the INTE command of CHARMM to predict the effects of mutations on PPIs. This command-line tool automates mutation analysis and interaction energy estimation, is applicable to different ligand types (protein, DNA, RNA, ion, small molecule) and provides various other features. The energy values yield absolute and normalized heat maps that allow rapid identification of stabilizing and destabilizing mutations. Our protocol forms the basis for automated programs that facilitate studies of binding-altering mutations in host-pathogen, protein-protein, and drug-target interactions.