Accelerating Molecular Docking using Machine Learning Methods.

Virtual screening (VS) is one of the well-established approaches in drug discovery which speeds up the search for a bioactive molecule and, reduces costs and efforts associated with experiments. VS helps to narrow down the search space of chemical space and allows selecting fewer and more probable candidate compounds for experimental testing. Docking calculations are one of the commonly used and highly appreciated structure-based drug discovery methods. Databases for chemical structures of small molecules have been growing rapidly. However, at the moment virtual screening of large libraries via docking is not very common. In this work, we aim to accelerate docking studies by predicting docking scores without explicitly performing docking calculations. We experimented with an attention based long short-term memory (LSTM) neural network for an efficient prediction of docking scores as well as other machine learning models such as XGBoost. By using docking scores of a small number of ligands we trained our models and predicted docking scores of a few million molecules. Specifically, we tested our approaches on 11 datasets that were produced from in-house drug discovery studies. On average, by training models using only 7000 molecules we predicted docking scores of approximately 3.8 million molecules with R2 (coefficient of determination) of 0.77 and Spearman rank correlation coefficient of 0.85. We designed the system with ease of use in mind. All the user needs to provide is a csv file containing SMILES and their respective docking scores, the system then outputs a model that the user can use for the prediction of docking score for a new molecule.

Depolymerization of actin filaments by Cucurbitacin I through binding G-actin.

Cucurbitacins have high economic value as they are a major source of food and have pharmacological properties. Cucurbitacin I (CuI) is a plant-derived natural tetracyclic triterpenoid compound that shows an anticancer effect via inhibiting the JAK2-STAT3 signaling pathway. The actin cytoskeleton is the most abundant protein in cells and regulates critical events through reorganization in cells. In this study, it is aimed at determining the direct effect of CuI on actin dynamics. The fluorescence profile of G-actin in the presence of CuI (1-200 nM) shifted to a higher temperature, suggesting that G-actin binds CuI and that G-actin-CuI is more thermally stable than the ligand-free form. CuI dose-dependently inhibited the polymerization of F-actin in vitro and disrupted actin filaments in endothelial cells. Docking and MD simulations suggested that CuI binds to the binding site formed by residues I136, I175, D154, and A138 that are at the interface of monomers in F-actin. The migration ability of cells treated with CuI for 24 h was significantly lower than the control group (p < .001). This study reveals the molecular mechanisms of CuI in the regulation of actin dynamics by binding G-actin. More importantly, this study indicates a novel role of CuI as an actin-targeting drug by binding directly to G-actin and may contribute to the mode of action of CuI on anticancer activities.

Optimization of CHARMM force field parameters for ryanodine receptor inhibitory drug dantrolene using FFTK and FFParam.

CONTEXT: Ryanodine receptors (RyRs) are large intracellular ligand-gated calcium release ion channels. Mutations in human RyR1 in combination with a volatile anesthetic or muscle relaxant are known to cause leaky RyRs resulting in malignant hyperthermia (MH). This has long been primarily treated with the RyR inhibitory drug dantrolene. Alternatives to dantrolene as a RyR inhibitor may be found through computer-aided drug design. Additionally, molecular dynamics (MD) studies of dantrolene interacting with RyRs may reveal its full mechanism of action. The availability of accurate force field parameters is important for the success of both. METHODS: In this study, force field parameters for dantrolene were obtained from the CHARMM General Force Field (CGenFF) program and optimized using the force field toolkit (FFTK) and FFParam programs. The obtained parameters were then validated by a comparison between calculated and experimental IR spectra and normal mode analysis, among other techniques.

The iron(III) and nickel(II) complexes with tetradentate thiosemicarbazones. Synthesis, experimental, theoretical characterization, and antiviral effect against SARS-CoV-2.

The discovery of new inhibitors that can be used in the treatment of viral diseases, including Covid-19, is an area open to research, and there is a need for innovative compounds with increased efficiency that provide inhibition by suppressing enzyme, and receptor mechanisms. The iron(III) and nickel(II) complexes were synthesized by template condensation of 4-methoxy-salicylaldehyde with S-methylthiosemicarbazone derivatives of 1,1,1-trifluoroacetylacetone (for Fe1) and methylacetoacetate (for Ni1). The complex structures having N2O2-chelating thiosemicarbazidato ligand were identified by analytical, spectroscopic, and X-ray crystallography results. Coordination environment of iron(III) center in complex Fe1 has a distorted square pyramidal geometry consisting of the N2O2 donor set and a chlorine atom, while that of Ni1 is square plane with the set. Inhibitory effect of Fe1 compound against SARS-CoV-2 virus specific 3C-like protease enzyme was investigated experimentally. It was determined that the highest inhibition concentration of Fe1 was 100 µM. Percent inhibition activity at this concentration was on average 30.62 ± 3.809%. Binding of both compounds to the 3C-like protease enzyme specific to the SARS-CoV-2 virus was analyzed using docking calculations. As a result of the docking calculation of Fe1, it has been observed that the compound has a binding energy of -7.4 kcal / mol to 3CL-like protease. It has been observed that the protein amino acids GLY143, THR26, and ASN142 contribute to the high binding affinity of the Fe1. The experimental and theoretical results obtained for the two complexes support each other.

Optimization of CHARMM force field parameters of a prostate cancer drug enzalutamide.

Enzalutamide is the most recent and potent clinically available drug molecule used to treat castration-resistant metastatic prostate cancer patients. In the presence of certain mutations on the Androgen receptor, enzalutamide fails to function as an antagonist. The molecular mechanism of this resistance has not been understood clearly. Molecular Dynamics (MD) simulations are a powerful tool to investigate such resistance mechanisms. Insights from MD simulations of receptor-ligand complex heavily depend on the quality of the parameters of the ligands. Thus, we developed and optimized CHARMM Force field parameters of enzalutamide. The initial assignment of parameters was done by using the CHARMM General Force Field (CGenFF) program. After that, using the Force Field Tool Kit (FFTK) partial charges, bond, angle and dihedral parameters of enzalutamide were optimized to reproduce quantum mechanics calculation results. For the validation of parameters, water-octanol partition coefficient, Infrared Spectra (IR) and normal mode analysis calculations were performed. In addition, a 300 ns-long MD simulation of enzalutamide was carried out to assess the ability of the optimized parameters in exploring torsional angle space.

Supramolecular Peptide Nanofiber Morphology Affects Mechanotransduction of Stem Cells.

Chirality and morphology are essential factors for protein function and interactions with other biomacromolecules. Extracellular matrix (ECM) proteins are also similar to other proteins in this sense; however, the complexity of the natural ECM makes it difficult to study these factors at the cellular level. The synthetic peptide nanomaterials harbor great promise in mimicking specific ECM molecules as model systems. In this work, we demonstrate that mechanosensory responses of stem cells are directly regulated by the chirality and morphology of ECM-mimetic peptide nanofibers with strictly controlled characteristics. Structural signals presented on l-amino acid containing cylindrical nanofibers (l-VV) favored the formation of integrin ß1-based focal adhesion complexes, which increased the osteogenic potential of stem cells through the activation of nuclear YAP. On the other hand, twisted ribbon-like nanofibers (l-FF and d-FF) guided the cells into round shapes and decreased the formation of focal adhesion complexes, which resulted in the confinement of YAP proteins in the cytosol and a corresponding decrease in osteogenic potential. Interestingly, the d-form of twisted-ribbon like nanofibers (d-FF) increased the chondrogenic potential of stem cells more than their l-form (l-FF). Our results provide new insights into the importance and relevance of morphology and chirality of nanomaterials in their interactions with cells and reveal that precise control over the chemical and physical properties of nanostructures can affect stem cell fate even without the incorporation of specific epitopes.

Hierarchical Self-Assembly of Histidine-Functionalized Peptide Amphiphiles into Supramolecular Chiral Nanostructures.

Controlling the hierarchical organization of self-assembling peptide amphiphiles into supramolecular nanostructures opens up the possibility of developing biocompatible functional supramolecular materials for various applications. In this study, we show that the hierarchical self-assembly of histidine- (His-) functionalized PAs containing d- or l-amino acids can be controlled by both solution pH and molecular chirality of the building blocks. An increase in solution pH resulted in the structural transition of the His-functionalized chiral PA assemblies from nanosheets to completely closed nanotubes through an enhanced hydrogen-bonding capacity and π-π stacking of imidazole ring. The effects of the stereochemistry and amino acid sequence of the PA backbone on the supramolecular organization were also analyzed by CD, TEM, SAXS, and molecular dynamics simulations. In addition, an investigation of chiral mixtures revealed the differences between the hydrogen-bonding capacities and noncovalent interactions of PAs with d- and l-amino acids.

Phylogenetic, structural, and functional characterization of AMT3;1, an ammonium transporter induced by mycorrhization among model grasses.

In the arbuscular mycorrhizal (AM) symbiosis, plants satisfy part of their nitrogen (N) requirement through the AM pathway. In sorghum, the ammonium transporters (AMT) AMT3;1, and to a lesser extent AMT4, are induced in cells containing developing arbuscules. Here, we have characterized orthologs of AMT3;1 and AMT4 in four other grasses in addition to sorghum. AMT3;1 and AMT4 orthologous genes are induced in AM roots, suggesting that in the common ancestor of these five plant species, both AMT3;1 and AMT4 were already present and upregulated upon AM colonization. An artificial microRNA approach was successfully used to downregulate either AMT3;1 or AMT4 in rice. Mycorrhizal root colonization and hyphal length density of knockdown plants were not affected at that time, indicating that the manipulation did not modify the establishment of the AM symbiosis and the interaction between both partners. However, expression of the fungal phosphate transporter FmPT was significantly reduced in knockdown plants, indicating a reduction of the nutrient fluxes from the AM fungus to the plant. The AMT3;1 knockdown plants (but not the AMT4 knockdown plants) were significantly less stimulated in growth by AM fungal colonization, and uptake of both 15N and 33P from the AM fungal network was reduced. This confirms that N and phosphorus nutrition through the mycorrhizal pathway are closely linked. But most importantly, it indicates that AMT3;1 is the prime plant transporter involved in the mycorrhizal ammonium transfer and that its function during uptake of N cannot be performed by AMT4.

Trisubstituted-Imidazoles Induce Apoptosis in Human Breast Cancer Cells by Targeting the Oncogenic PI3K/Akt/mTOR Signaling Pathway.

Overactivation of PI3K/Akt/mTOR is linked with carcinogenesis and serves a potential molecular therapeutic target in treatment of various cancers. Herein, we report the synthesis of trisubstituted-imidazoles and identified 2-chloro-3-(4, 5-diphenyl-1H-imidazol-2-yl) pyridine (CIP) as lead cytotoxic agent. Naïve Base classifier model of in silico target prediction revealed that CIP targets RAC-beta serine/threonine-protein kinase which comprises the Akt. Furthermore, CIP downregulated the phosphorylation of Akt, PDK and mTOR proteins and decreased expression of cyclin D1, Bcl-2, survivin, VEGF, procaspase-3 and increased cleavage of PARP. In addition, CIP significantly downregulated the CXCL12 induced motility of breast cancer cells and molecular docking calculations revealed that all compounds bind to Akt2 kinase with high docking scores compared to the library of previously reported Akt2 inhibitors. In summary, we report the synthesis and biological evaluation of imidazoles that induce apoptosis in breast cancer cells by negatively regulating PI3K/Akt/mTOR signaling pathway.

Catch-bond mechanism of the bacterial adhesin FimH.

Ligand-receptor interactions that are reinforced by mechanical stress, so-called catch-bonds, play a major role in cell-cell adhesion. They critically contribute to widespread urinary tract infections by pathogenic Escherichia coli strains. These pathogens attach to host epithelia via the adhesin FimH, a two-domain protein at the tip of type I pili recognizing terminal mannoses on epithelial glycoproteins. Here we establish peptide-complemented FimH as a model system for fimbrial FimH function. We reveal a three-state mechanism of FimH catch-bond formation based on crystal structures of all states, kinetic analysis of ligand interaction and molecular dynamics simulations. In the absence of tensile force, the FimH pilin domain allosterically accelerates spontaneous ligand dissociation from the FimH lectin domain by 100,000-fold, resulting in weak affinity. Separation of the FimH domains under stress abolishes allosteric interplay and increases the affinity of the lectin domain. Cell tracking demonstrates that rapid ligand dissociation from FimH supports motility of piliated E. coli on mannosylated surfaces in the absence of shear force.

Mechanism of NH4(+) Recruitment and NH3 Transport in Rh Proteins.

In human cells, membrane proteins of the rhesus (Rh) family excrete ammonium and play a role in pH regulation. Based on high-resolution structures, Rh proteins are generally understood to act as NH3 channels. Given that cell membranes are permeable to gases like NH3, the role of such proteins remains a paradox. Using molecular and quantum mechanical calculations, we show that a crystallographically identified site in the RhCG pore actually recruits NH4(+), which is found in higher concentration and binds with higher affinity than NH3, increasing the efficiency of the transport mechanism. A proton is transferred from NH4(+) to a signature histidine (the only moiety thermodynamically likely to accept a proton) followed by the diffusion of NH3 down the pore. The excess proton is circulated back to the extracellular vestibule through a hydrogen bond network, which involves a highly conserved and functionally important aspartic acid, resulting in the net transport of NH3.

A PDI-catalyzed thiol-disulfide switch regulates the production of hydrogen peroxide by human Ero1.

Oxidative folding in the endoplasmic reticulum (ER) involves ER oxidoreductin 1 (Ero1)-mediated disulfide formation in protein disulfide isomerase (PDI). In this process, Ero1 consumes oxygen (O2) and releases hydrogen peroxide (H2O2), but none of the published Ero1 crystal structures reveal any potential pathway for entry and exit of these reactants. We report that additional mutation of the Cys(208)-Cys(241) disulfide in hyperactive Ero1α (Ero1α-C104A/C131A) potentiates H2O2 production, ER oxidation, and cell toxicity. This disulfide clamps two helices that seal the flavin cofactor where O2 is reduced to H2O2. Through its carboxyterminal active site, PDI unlocks this seal by forming a Cys(208)/Cys(241)-dependent mixed-disulfide complex with Ero1α. The H2O2-detoxifying glutathione peroxidase 8 also binds to the Cys(208)/Cys(241) loop region. Supported by O2 diffusion simulations, these data describe the first enzymatically controlled O2 access into a flavoprotein active site, provide molecular-level understanding of Ero1α regulation and H2O2 production/detoxification, and establish the deleterious consequences of constitutive Ero1 activity.

Functional asymmetry within the Sec61p translocon.

The Sec61 translocon forms a pore to translocate polypeptide sequences across the membrane and offers a lateral gate for membrane integration of hydrophobic (H) segments. A central constriction of six apolar residues has been shown to form a seal, but also to determine the hydrophobicity threshold for membrane integration: Mutation of these residues in yeast Sec61p to glycines, serines, aspartates, or lysines lowered the hydrophobicity required for integration; mutation to alanines increased it. Whereas four leucines distributed in an oligo-alanine H segment were sufficient for 50% integration, we now find four leucines in the N-terminal half of the H segment to produce significantly more integration than in the C-terminal half, suggesting functional asymmetry within the translocon. Scanning a cluster of three leucines through an oligo-alanine H segment showed high integration levels, except around the position matching that of the hydrophobic constriction in the pore where integration was strongly reduced. Both asymmetry and the position effect of H-segment integration disappeared upon mutation of the constriction residues to glycines or serines, demonstrating that hydrophobicity at this position within the translocon is responsible for the phenomenon. Asymmetry was largely retained, however, when constriction residues were replaced by alanines. These results reflect on the integration mechanism of transmembrane domains and show that membrane insertion of H segments strongly depends not only on their intrinsic hydrophobicity but also on the local conditions in the translocon interior. Thus, the contribution of hydrophobic residues in the H segment is not simply additive and displays cooperativeness depending on their relative position.

Different hydration patterns in the pores of AmtB and RhCG could determine their transport mechanisms.

The ammonium transporters of the Amt/Rh family facilitate the diffusion of ammonium across cellular membranes. Functional data suggest that Amt proteins, notably found in plants, transport the ammonium ion (NH4(+)), whereas human Rhesus (Rh) proteins transport ammonia (NH3). Comparison between the X-ray structures of the prokaryotic AmtB, assumed to be representative of Amt proteins, and the human RhCG reveals important differences at the level of their pore. Despite these important functional and structural differences between Amt and Rh proteins, studies of the AmtB transporter have led to the suggestion that proteins of both subfamilies work according to the same mechanism and transport ammonia. We performed molecular dynamics simulations of the AmtB and RhCG proteins under different water and ammonia occupancy states of their pore. Free energy calculations suggest that the probability of finding NH3 molecules in the pore of AmtB is negligible in comparison to finding water. The presence of water in the pore of AmtB could support the transport of proton. The pore lumen of RhCG is found to be more hydrophobic due to the presence of a phenylalanine conserved among Rh proteins. Simulations of RhCG also reveal that the signature histidine dyad is occasionally exposed to the extracellular bulk, which is never observed in AmtB. These different hydration patterns are consistent with the idea that Amt and Rh proteins are not functionally equivalent and that permeation takes place according to two distinct mechanisms.

Ammonium transporters achieve charge transfer by fragmenting their substrate.

Proteins of the Amt/MEP family facilitate ammonium transport across the membranes of plants, fungi, and bacteria and are essential for growth in nitrogen-poor environments. Some are known to facilitate the diffusion of the neutral NH(3), while others, notably in plants, transport the positively charged NH(4)(+). On the basis of the structural data for AmtB from Escherichia coli , we illustrate the mechanism by which proteins from the Amt family can sustain electrogenic transport. Free energy calculations show that NH(4)(+) is stable in the AmtB pore, reaching a binding site from which it can spontaneously transfer a proton to a pore-lining histidine residue (His168). The substrate diffuses down the pore in the form of NH(3), while the excess proton is cotransported through a highly conserved hydrogen-bonded His168-His318 pair. This constitutes a novel permeation mechanism that confers to the histidine dyad an essential mechanistic role that was so far unknown.

Determination of pair-wise inter-residue interaction forces from folding pathways and their implementation in coarse-grained folding prediction.

Interaction forces among residue pairs are determined from optimum folding pathways along which a protein represented as a coarse-grained chain of alpha-carbons goes from different initial configurations to a known native state. A dynamic optimization approach is employed that uses the coarse-grained model to compute the optimal folding pathways. The pair-wise interaction forces obtained in this manner are incorporated into the coarse-grained model which is then simulated to fold the protein from a new set of initial configurations in a predictive way. We show that the folding pathways predicted in this manner are near-optimal. We applied the technique to the secondary structures: helix and beta-sheet.

Optimum folding pathways for growing protein chains.

The folding of a protein is studied as it grows residue by residue from the N-terminus and enters an environment that stabilizes the folded state. This mode of folding of a growing chain is different from refolding where the full chain folds from a disordered initial configuration to the native state. We propose a sequential dynamic optimization method that computes the evolution of optimum folding pathways as amino acid residues are added to the peptide chain one by one. The dynamic optimization formulation is deterministic and uses Newton's equations of motion and a Go-type potential that establishes the native contacts and excluded volume effects. The method predicts the optimal energy-minimizing path among all the alternative feasible pathways. As two examples, the folding of the chicken villin headpiece, a 36-residue protein, and chymotrypsin inhibitor 2 (CI2), a 64-residue protein, are studied. Results on the villin headpiece show significant differences from the refolding of the same chain studied previously. Results on CI2 mostly agree with the results of refolding experiments and computational work.