The Art and Science of Molecular Docking.

Molecular docking has become an essential part of a structural biologist's and medicinal chemist's toolkits. Given a chemical compound and the three-dimensional structure of a molecular target-for example, a protein-docking methods fit the compound into the target, predicting the compound's bound structure and binding energy. Docking can be used to discover novel ligands for a target by screening large virtual compound libraries. Docking can also provide a useful starting point for structure-based ligand optimization or for investigating a ligand's mechanism of action. Advances in computational methods, including both physics-based and machine learning approaches, as well as in complementary experimental techniques, are making docking an even more powerful tool. We review how docking works and how it can drive drug discovery and biological research. We also describe its current limitations and ongoing efforts to overcome them.

Structure-based discovery of CFTR potentiators and inhibitors.

The cystic fibrosis transmembrane conductance regulator (CFTR) is a crucial ion channel whose loss of function leads to cystic fibrosis, whereas its hyperactivation leads to secretory diarrhea. Small molecules that improve CFTR folding (correctors) or function (potentiators) are clinically available. However, the only potentiator, ivacaftor, has suboptimal pharmacokinetics and inhibitors have yet to be clinically developed. Here, we combine molecular docking, electrophysiology, cryo-EM, and medicinal chemistry to identify CFTR modulators. We docked ∼155 million molecules into the potentiator site on CFTR, synthesized 53 test ligands, and used structure-based optimization to identify candidate modulators. This approach uncovered mid-nanomolar potentiators, as well as inhibitors, that bind to the same allosteric site. These molecules represent potential leads for the development of more effective drugs for cystic fibrosis and secretory diarrhea, demonstrating the feasibility of large-scale docking for ion channel drug discovery.

Structure-based discovery of conformationally selective inhibitors of the serotonin transporter.

The serotonin transporter (SERT) removes synaptic serotonin and is the target of anti-depressant drugs. SERT adopts three conformations: outward-open, occluded, and inward-open. All known inhibitors target the outward-open state except ibogaine, which has unusual anti-depressant and substance-withdrawal effects, and stabilizes the inward-open conformation. Unfortunately, ibogaine's promiscuity and cardiotoxicity limit the understanding of inward-open state ligands. We docked over 200 million small molecules against the inward-open state of the SERT. Thirty-six top-ranking compounds were synthesized, and thirteen inhibited; further structure-based optimization led to the selection of two potent (low nanomolar) inhibitors. These stabilized an outward-closed state of the SERT with little activity against common off-targets. A cryo-EM structure of one of these bound to the SERT confirmed the predicted geometry. In mouse behavioral assays, both compounds had anxiolytic- and anti-depressant-like activity, with potencies up to 200-fold better than fluoxetine (Prozac), and one substantially reversed morphine withdrawal effects.

An infectivity-enhancing site on the SARS-CoV-2 spike protein targeted by antibodies.

Antibodies against the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein prevent SARS-CoV-2 infection. However, the effects of antibodies against other spike protein domains are largely unknown. Here, we screened a series of anti-spike monoclonal antibodies from coronavirus disease 2019 (COVID-19) patients and found that some of antibodies against the N-terminal domain (NTD) induced the open conformation of RBD and thus enhanced the binding capacity of the spike protein to ACE2 and infectivity of SARS-CoV-2. Mutational analysis revealed that all of the infectivity-enhancing antibodies recognized a specific site on the NTD. Structural analysis demonstrated that all infectivity-enhancing antibodies bound to NTD in a similar manner. The antibodies against this infectivity-enhancing site were detected at high levels in severe patients. Moreover, we identified antibodies against the infectivity-enhancing site in uninfected donors, albeit at a lower frequency. These findings demonstrate that not only neutralizing antibodies but also enhancing antibodies are produced during SARS-CoV-2 infection.

ATG8-Binding UIM Proteins Define a New Class of Autophagy Adaptors and Receptors.

During autophagy, vesicle dynamics and cargo recruitment are driven by numerous adaptors and receptors that become tethered to the phagophore through interactions with lipidated ATG8/LC3 decorating the expanding membrane. Most currently described ATG8-binding proteins exploit a well-defined ATG8-interacting motif (AIM, or LC3-interacting region [LIR]) that contacts a hydrophobic patch on ATG8 known as the LIR/AIM docking site (LDS). Here we describe a new class of ATG8 interactors that exploit ubiquitin-interacting motif (UIM)-like sequences for high-affinity binding to an alternative ATG8 interaction site. Assays with candidate UIM-containing proteins together with unbiased screens identified a large collection of UIM-based ATG8 interactors in plants, yeast, and humans. Analysis of a subset also harboring ubiquitin regulatory X (UBX) domains revealed a role for UIM-directed autophagy in clearing non-functional CDC48/p97 complexes, including some impaired in human disease. With this new class of adaptors and receptors, we greatly extend the reach of selective autophagy and identify new factors regulating autophagic vesicle dynamics.

Irisin acts through its integrin receptor in a two-step process involving extracellular Hsp90α.

Exercise benefits the human body in many ways. Irisin is secreted by muscle, increased with exercise, and conveys physiological benefits, including improved cognition and resistance to neurodegeneration. Irisin acts via αV integrins; however, a mechanistic understanding of how small polypeptides like irisin can signal through integrins is poorly understood. Using mass spectrometry and cryo-EM, we demonstrate that the extracellular heat shock protein 90α (eHsp90α) is secreted by muscle with exercise and activates integrin αVß5. This allows for high-affinity irisin binding and signaling through an Hsp90α/αV/ß5 complex. By including hydrogen/deuterium exchange data, we generate and experimentally validate a 2.98 Å RMSD irisin/αVß5 complex docking model. Irisin binds very tightly to an alternative interface on αVß5 distinct from that used by known ligands. These data elucidate a non-canonical mechanism by which a small polypeptide hormone like irisin can function through an integrin receptor.

PEX5 translocation into and out of peroxisomes drives matrix protein import.

Peroxisomes are ubiquitous organelles whose dysfunction causes fatal human diseases. Most peroxisomal enzymes are imported from the cytosol by the receptor PEX5, which interacts with a docking complex in the peroxisomal membrane and then returns to the cytosol after monoubiquitination by a membrane-embedded ubiquitin ligase. The mechanism by which PEX5 shuttles between cytosol and peroxisomes and releases cargo inside the lumen is unclear. Here, we use Xenopus egg extract to demonstrate that PEX5 accompanies cargo completely into the lumen, utilizing WxxxF/Y motifs near its N terminus that bind a lumenal domain of the docking complex. PEX5 recycling is initiated by an amphipathic helix that binds to the lumenal side of the ubiquitin ligase. The N terminus then emerges in the cytosol for monoubiquitination. Finally, PEX5 is extracted from the lumen, resulting in the unfolding of the receptor and cargo release. Our results reveal the unique mechanism by which PEX5 ferries proteins into peroxisomes.

Recent advances in predicting and modeling protein-protein interactions.

Protein-protein interactions (PPIs) drive biological processes, and disruption of PPIs can cause disease. With recent breakthroughs in structure prediction and a deluge of genomic sequence data, computational methods to predict PPIs and model spatial structures of protein complexes are now approaching the accuracy of experimental approaches for permanent interactions and show promise for elucidating transient interactions. As we describe here, the key to this success is rich evolutionary information deciphered from thousands of homologous sequences that coevolve in interacting partners. This covariation signal, revealed by sophisticated statistical and machine learning (ML) algorithms, predicts physiological interactions. Accurate artificial intelligence (AI)-based modeling of protein structures promises to provide accurate 3D models of PPIs at a proteome-wide scale.

Localisation and function of key axonemal microtubule inner proteins and dynein docking complex members reveal extensive diversity among vertebrate motile cilia.

Vertebrate motile cilia are classified as (9+2) or (9+0), based on the presence or absence of the central pair apparatus, respectively. Cryogenic electron microscopy analyses of (9+2) cilia have uncovered an elaborate axonemal protein composition. The extent to which these features are conserved in (9+0) cilia remains unclear. CFAP53, a key axonemal filamentous microtubule inner protein (fMIP) and a centriolar satellites component, is essential for motility of (9+0), but not (9+2) cilia. Here, we show that in (9+2) cilia, CFAP53 functions redundantly with a paralogous fMIP, MNS1. MNS1 localises to ciliary axonemes, and combined loss of both proteins in zebrafish and mice caused severe outer dynein arm loss from (9+2) cilia, significantly affecting their motility. Using immunoprecipitation, we demonstrate that, whereas MNS1 can associate with itself and CFAP53, CFAP53 is unable to self-associate. We also show that additional axonemal dynein-interacting proteins, two outer dynein arm docking (ODAD) complex members, show differential localisation between types of motile cilia. Together, our findings clarify how paralogous fMIPs, CFAP53 and MNS1, function in regulating (9+2) versus (9+0) cilia motility, and further emphasise extensive structural diversity among these organelles.

Cyclin-Specific Docking Mechanisms Reveal the Complexity of M-CDK Function in the Cell Cycle.

Cyclin-dependent kinases (CDKs) coordinate hundreds of molecular events during the cell cycle. Multiple cyclins are involved, but the global role of cyclin-specific phosphorylation has remained unsolved. We uncovered a cyclin docking motif, LxF, that mediates binding of replication factor Cdc6 to mitotic cyclin. This interaction leads to phospho-adaptor Cks1-mediated inhibition of M-CDK to facilitate Cdc6 accumulation and sequestration in mitosis. The LxF motif and Cks1 also mediate the mutual inhibition between M-CDK and the tyrosine kinase Swe1. Additionally, the LxF motif is critical for targeting M-CDK to phosphorylate several mitotic regulators; for example, Spo12 is targeted via LxF to release the phosphatase Cdc14. The results complete the full set of G1, S, and M-CDK docking mechanisms and outline the unified role of cyclin specificity and CDK activity thresholds. Cooperation of cyclin and Cks1 docking creates a variety of CDK thresholds and switching orders, including combinations of last in, first out (LIFO) and first in, first out (FIFO) ordering.

Cyclin D-Cdk4,6 Drives Cell-Cycle Progression via the Retinoblastoma Protein's C-Terminal Helix.

The cyclin-dependent kinases Cdk4 and Cdk6 form complexes with D-type cyclins to drive cell proliferation. A well-known target of cyclin D-Cdk4,6 is the retinoblastoma protein Rb, which inhibits cell-cycle progression until its inactivation by phosphorylation. However, the role of Rb phosphorylation by cyclin D-Cdk4,6 in cell-cycle progression is unclear because Rb can be phosphorylated by other cyclin-Cdks, and cyclin D-Cdk4,6 has other targets involved in cell division. Here, we show that cyclin D-Cdk4,6 docks one side of an alpha-helix in the Rb C terminus, which is not recognized by cyclins E, A, and B. This helix-based docking mechanism is shared by the p107 and p130 Rb-family members across metazoans. Mutation of the Rb C-terminal helix prevents its phosphorylation, promotes G1 arrest, and enhances Rb's tumor suppressive function. Our work conclusively demonstrates that the cyclin D-Rb interaction drives cell division and expands the diversity of known cyclin-based protein docking mechanisms.

A conserved peptide-binding pocket in HyNaC/ASIC ion channels.

The only known peptide-gated ion channels-FaNaCs/WaNaCs and HyNaCs-belong to different clades of the DEG/ENaC family. FaNaCs are activated by the short neuropeptide FMRFamide, and HyNaCs by Hydra RFamides, which are not evolutionarily related to FMRFamide. The FMRFamide-binding site in FaNaCs was recently identified in a cleft atop the large extracellular domain. However, this cleft is not conserved in HyNaCs. Here, we combined molecular modeling and site-directed mutagenesis and identified a putative binding pocket for Hydra-RFamides in the extracellular domain of the heterotrimeric HyNaC2/3/5. This pocket localizes to only one of the three subunit interfaces, indicating that this trimeric ion channel binds a single peptide ligand. We engineered an unnatural amino acid at the putative binding pocket entrance, which allowed covalent tethering of Hydra RFamide to the channel, thereby trapping the channel in an open conformation. The identified pocket localizes to the same region as the acidic pocket of acid-sensing ion channels (ASICs), which binds peptide ligands. The pocket in HyNaCs is less acidic, and both electrostatic and hydrophobic interactions contribute to peptide binding. Collectively, our results reveal a conserved ligand-binding pocket in HyNaCs and ASICs and indicate independent evolution of peptide-binding cavities in the two subgroups of peptide-gated ion channels.

ABAG-docking benchmark: a non-redundant structure benchmark dataset for antibody-antigen computational docking.

Accurate prediction of antibody-antigen complex structures is pivotal in drug discovery, vaccine design and disease treatment and can facilitate the development of more effective therapies and diagnostics. In this work, we first review the antibody-antigen docking (ABAG-docking) datasets. Then, we present the creation and characterization of a comprehensive benchmark dataset of antibody-antigen complexes. We categorize the dataset based on docking difficulty, interface properties and structural characteristics, to provide a diverse set of cases for rigorous evaluation. Compared with Docking Benchmark 5.5, we have added 112 cases, including 14 single-domain antibody (sdAb) cases and 98 monoclonal antibody (mAb) cases, and also increased the proportion of Difficult cases. Our dataset contains diverse cases, including human/humanized antibodies, sdAbs, rodent antibodies and other types, opening the door to better algorithm development. Furthermore, we provide details on the process of building the benchmark dataset and introduce a pipeline for periodic updates to keep it up to date. We also utilize multiple complex prediction methods including ZDOCK, ClusPro, HDOCK and AlphaFold-Multimer for testing and analyzing this dataset. This benchmark serves as a valuable resource for evaluating and advancing docking computational methods in the analysis of antibody-antigen interaction, enabling researchers to develop more accurate and effective tools for predicting and designing antibody-antigen complexes. The non-redundant ABAG-docking structure benchmark dataset is available at https://github.com/Zhaonan99/Antibody-antigen-complex-structure-benchmark-dataset.

PGBind: pocket-guided explicit attention learning for protein-ligand docking.

As more and more protein structures are discovered, blind protein-ligand docking will play an important role in drug discovery because it can predict protein-ligand complex conformation without pocket information on the target proteins. Recently, deep learning-based methods have made significant advancements in blind protein-ligand docking, but their protein features are suboptimal because they do not fully consider the difference between potential pocket regions and non-pocket regions in protein feature extraction. In this work, we propose a pocket-guided strategy for guiding the ligand to dock to potential docking regions on a protein. To this end, we design a plug-and-play module to enhance the protein features, which can be directly incorporated into existing deep learning-based blind docking methods. The proposed module first estimates potential pocket regions on the target protein and then leverages a pocket-guided attention mechanism to enhance the protein features. Experiments are conducted on integrating our method with EquiBind and FABind, and the results show that their blind-docking performances are both significantly improved and new start-of-the-art performance is achieved by integration with FABind.

RmsdXNA: RMSD prediction of nucleic acid-ligand docking poses using machine-learning method.

Small molecule drugs can be used to target nucleic acids (NA) to regulate biological processes. Computational modeling methods, such as molecular docking or scoring functions, are commonly employed to facilitate drug design. However, the accuracy of the scoring function in predicting the closest-to-native docking pose is often suboptimal. To overcome this problem, a machine learning model, RmsdXNA, was developed to predict the root-mean-square-deviation (RMSD) of ligand docking poses in NA complexes. The versatility of RmsdXNA has been demonstrated by its successful application to various complexes involving different types of NA receptors and ligands, including metal complexes and short peptides. The predicted RMSD by RmsdXNA was strongly correlated with the actual RMSD of the docked poses. RmsdXNA also outperformed the rDock scoring function in ranking and identifying closest-to-native docking poses across different structural groups and on the testing dataset. Using experimental validated results conducted on polyadenylated nuclear element for nuclear expression triplex, RmsdXNA demonstrated better screening power for the RNA-small molecule complex compared to rDock. Molecular dynamics simulations were subsequently employed to validate the binding of top-scoring ligand candidates selected by RmsdXNA and rDock on MALAT1. The results showed that RmsdXNA has a higher success rate in identifying promising ligands that can bind well to the receptor. The development of an accurate docking score for a NA-ligand complex can aid in drug discovery and development advancements. The code to use RmsdXNA is available at the GitHub repository https://github.com/laiheng001/RmsdXNA.

GSScore: a novel Graphormer-based shell-like scoring method for protein-ligand docking.

Protein-ligand interactions (PLIs) are essential for cellular activities and drug discovery. But due to the complexity and high cost of experimental methods, there is a great demand for computational approaches to recognize PLI patterns, such as protein-ligand docking. In recent years, more and more models based on machine learning have been developed to directly predict the root mean square deviation (RMSD) of a ligand docking pose with reference to its native binding pose. However, new scoring methods are pressingly needed in methodology for more accurate RMSD prediction. We present a new deep learning-based scoring method for RMSD prediction of protein-ligand docking poses based on a Graphormer method and Shell-like graph architecture, named GSScore. To recognize near-native conformations from a set of poses, GSScore takes atoms as nodes and then establishes the docking interface of protein-ligand into multiple bipartite graphs within different shell ranges. Benefiting from the Graphormer and Shell-like graph architecture, GSScore can effectively capture the subtle differences between energetically favorable near-native conformations and unfavorable non-native poses without extra information. GSScore was extensively evaluated on diverse test sets including a subset of PDBBind version 2019, CASF2016 as well as DUD-E, and obtained significant improvements over existing methods in terms of RMSE, $R$ (Pearson correlation coefficient), Spearman correlation coefficient and Docking power.

Enhancing cryo-EM structure prediction with DeepTracer and AlphaFold2 integration.

Understanding the protein structures is invaluable in various biomedical applications, such as vaccine development. Protein structure model building from experimental electron density maps is a time-consuming and labor-intensive task. To address the challenge, machine learning approaches have been proposed to automate this process. Currently, the majority of the experimental maps in the database lack atomic resolution features, making it challenging for machine learning-based methods to precisely determine protein structures from cryogenic electron microscopy density maps. On the other hand, protein structure prediction methods, such as AlphaFold2, leverage evolutionary information from protein sequences and have recently achieved groundbreaking accuracy. However, these methods often require manual refinement, which is labor intensive and time consuming. In this study, we present DeepTracer-Refine, an automated method that refines AlphaFold predicted structures by aligning them to DeepTracers modeled structure. Our method was evaluated on 39 multi-domain proteins and we improved the average residue coverage from 78.2 to 90.0% and average local Distance Difference Test score from 0.67 to 0.71. We also compared DeepTracer-Refine with Phenixs AlphaFold refinement and demonstrated that our method not only performs better when the initial AlphaFold model is less precise but also surpasses Phenix in run-time performance.

Novel systems biology experimental pipeline reveals matairesinol's antimetastatic potential in prostate cancer: an integrated approach of network pharmacology, bioinformatics, and experimental validation.

Matairesinol (MAT), a plant lignan renowned for its anticancer properties in hormone-sensitive cancers like breast and prostate cancers, presents a promising yet underexplored avenue in the treatment of metastatic prostate cancer (mPC). To elucidate its specific therapeutic targets and mechanisms, our study adopted an integrative approach, amalgamating network pharmacology (NP), bioinformatics, GeneMANIA-based functional association (GMFA), and experimental validation. By mining online databases, we identified 27 common targets of mPC and MAT, constructing a MAT-mPC protein-protein interaction network via STRING and pinpointing 11 hub targets such as EGFR, AKT1, ERBB2, MET, IGF1, CASP3, HSP90AA1, HIF1A, MMP2, HGF, and MMP9 with CytoHuba. Utilizing DAVID, Gene Ontology (GO) analysis highlighted metastasis-related processes such as epithelial-mesenchymal transition, positive regulation of cell migration, and key Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, including cancer, prostate cancer, PI3K-Akt, and MAPK signaling, while the web resources such as UALCAN and GEPIA2 affirmed the clinical significance of the top 11 hub targets in mPC patient survival analysis and gene expression patterns. Our innovative GMFA enrichment method further enriched network pharmacology findings. Molecular docking analyses demonstrated substantial interactions between MAT and 11 hub targets. Simulation studies confirmed the stable interactions of MAT with selected targets. Experimental validation in PC3 cells, employing quantitative real-time reverse-transcription PCR and various cell-based assays, corroborated MAT's antimetastatic effects on mPC. Thus, this exhaustive NP analysis, complemented by GMFA, molecular docking, molecular dynamics simulations, and experimental validations, underscores MAT's multifaceted role in targeting mPC through diverse therapeutic avenues. Nevertheless, comprehensive in vitro validation is imperative to solidify these findings.

Linear motif specificity in signaling through p38α and ERK2 mitogen-activated protein kinases.

Mitogen-activated protein kinase (MAPK) cascades are essential for eukaryotic cells to integrate and respond to diverse stimuli. Maintaining specificity in signaling through MAPK networks is key to coupling distinct inputs to appropriate cellular responses. Docking sites-short linear motifs found in MAPK substrates, regulators, and scaffolds-can promote signaling specificity through selective interactions, but how they do so remains unresolved. Here, we screened a proteomic library for sequences interacting with the MAPKs extracellular signal-regulated kinase 2 (ERK2) and p38α, identifying selective and promiscuous docking motifs. Sequences specific for p38α had high net charge and lysine content, and selective binding depended on a pair of acidic residues unique to the p38α docking interface. Finally, we validated a set of full-length proteins harboring docking sites selected in our screens to be authentic MAPK interactors and substrates. This study identifies features that help define MAPK signaling networks and explains how specific docking motifs promote signaling integrity.

Protein Amyloid Cofactors: Charged Side-Chain Arrays Meet Their Match?

Recent advances in high-resolution structural studies of protein amyloids have revealed parallel in-register cross-ß-sheets with periodic arrays of closely spaced identical residues. What do these structures tell us about the mechanisms of action of common amyloid-promoting factors, such as heparan sulfate (HS), nucleic acids, polyphosphates, anionic phospholipids, and acidic pH?