Determination of binding points of methylene blue and cationic phenoxazine dyes on human butyrylcholinesterase.

In this study, the binding points of MethB and two structurally-related cationic phenoxazine dyes [meldola blue (MB) and nile blue (NB)] to human butyrylcholinesterase (BChE) were investigated by molecular docking and site directed mutagenesis. The comparative inhibitory effects of MethB, MB and NB on recombinant wild type BChE and six human BChE mutants were spectrophotometrically studied. Kinetic analyses yielded the following information: MethB and MB were found to cause nonlinear inhibition of all recombinant BChEs except Y332A, compatible with a multi-site binding model. On the other hand, MethB and MB caused linear mixed inhibition of Y332A mutant, compatible with a single binding mode. Comparing the inhibitory effects in aspect of Ki values with recombinant wild type BChE (Ki=0.042 µM), MethB was found to be ∼30, 80 and 270-fold less effective as an inhibitor of Y332A, F329A and T120F, respectively. NB caused nonlinear inhibition of all recombinant BChEs. The inhibitory effect of NB on Y332A mutant was ∼370-fold lower, compared to recombinant wild type BChE (Ki=0.006 µM). Considering both kinetic and molecular docking results together, it was concluded that threonine 120, phenylalanine 329 and tyrosine 332 are critical amino acids in binding of cationic phenoxazine/phenothiazine structured ligands to human BChE.

Synthesis of PDE IV inhibitors. First asymmetric synthesis of two of GlaxoSmithKline's highly potent Rolipram analogues.

Asymmetric syntheses of two of GlaxoSmithKline's highly potent phosphodiesterase IV inhibitors CMPI 1 and CMPO 2 have been accomplished from nitroethane and simple precursors in 8 and 7 steps, respectively. The suggested synthetic strategy involves as a key stage the silylation of enantiopure six-membered cyclic nitronates. In vitro studies of PDE IVB1 inhibition revealed a significant difference in the activity of CMPI 1 and CMPO 2 enantiomers.

Predicting ligand binding modes from neural networks trained on protein-ligand interaction fingerprints.

We herewith present a novel approach to predict protein-ligand binding modes from the single two-dimensional structure of the ligand. Known protein-ligand X-ray structures were converted into binary bit strings encoding protein-ligand interactions. An artificial neural network was then set up to first learn and then predict protein-ligand interaction fingerprints from simple ligand descriptors. Specific models were constructed for three targets (CDK2, p38-α, HSP90-α) and 146 ligands for which protein-ligand X-ray structures are available. These models were able to predict protein-ligand interaction fingerprints and to discriminate important features from minor interactions. Predicted interaction fingerprints were successfully used as descriptors to discriminate true ligands from decoys by virtual screening. In some but not all cases, the predicted interaction fingerprints furthermore enable to efficiently rerank cross-docking poses and prioritize the best possible docking solutions.

Development and implementation of an enterprise-wide predictive model for early absorption, distribution, metabolism and excretion properties.

Background: Accurate prediction of absorption, distribution, metabolism and excretion (ADME) properties can facilitate the identification of promising drug candidates. Methodology & Results: The authors present the Janssen generic Target Product Profile (gTPP) model, which predicts 18 early ADME properties, employs a graph convolutional neural network algorithm and was trained on between 1000-10,000 internal data points per predicted parameter. gTPP demonstrated stronger predictive power than pretrained commercial ADME models and automatic model builders. Through a novel logging method, the authors report gTPP usage for more than 200 Janssen drug discovery scientists. Conclusion: The investigators successfully enabled the rapid and systematic implementation of predictive ML tools across a drug discovery pipeline in all therapeutic areas. This experience provides useful guidance for other large-scale AI/ML deployment efforts.

Industry-scale application and evaluation of deep learning for drug target prediction.

Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, the quality of public data might be different than that of industry data due to different labs reporting measurements, different measurement techniques, fewer samples and less diverse and specialized assays. As part of a European funded project (ExCAPE), that brought together expertise from pharmaceutical industry, machine learning, and high-performance computing, we investigated how well machine learning models obtained from public data can be transferred to internal pharmaceutical industry data. Our results show that machine learning models trained on public data can indeed maintain their predictive power to a large degree when applied to industry data. Moreover, we observed that deep learning derived machine learning models outperformed comparable models, which were trained by other machine learning algorithms, when applied to internal pharmaceutical company datasets. To our knowledge, this is the first large-scale study evaluating the potential of machine learning and especially deep learning directly at the level of industry-scale settings and moreover investigating the transferability of publicly learned target prediction models towards industrial bioactivity prediction pipelines.

Repurposing High-Throughput Image Assays Enables Biological Activity Prediction for Drug Discovery.

In both academia and the pharmaceutical industry, large-scale assays for drug discovery are expensive and often impractical, particularly for the increasingly important physiologically relevant model systems that require primary cells, organoids, whole organisms, or expensive or rare reagents. We hypothesized that data from a single high-throughput imaging assay can be repurposed to predict the biological activity of compounds in other assays, even those targeting alternate pathways or biological processes. Indeed, quantitative information extracted from a three-channel microscopy-based screen for glucocorticoid receptor translocation was able to predict assay-specific biological activity in two ongoing drug discovery projects. In these projects, repurposing increased hit rates by 50- to 250-fold over that of the initial project assays while increasing the chemical structure diversity of the hits. Our results suggest that data from high-content screens are a rich source of information that can be used to predict and replace customized biological assays.

Discovery of N-(Pyridin-4-yl)-1,5-naphthyridin-2-amines as Potential Tau Pathology PET Tracers for Alzheimer's Disease.

A mini-HTS on 4000 compounds selected using 2D fragment-based similarity and 3D pharmacophoric and shape similarity to known selective tau aggregate binders identified N-(6-methylpyridin-2-yl)quinolin-2-amine 10 as a novel potent binder to human AD aggregated tau with modest selectivity versus aggregated ß-amyloid (Aß). Initial medicinal chemistry efforts identified key elements for potency and selectivity, as well as suitable positions for radiofluorination, leading to a first generation of fluoroalkyl-substituted quinoline tau binding ligands with suboptimal physicochemical properties. Further optimization toward a more optimal pharmacokinetic profile led to the discovery of 1,5-naphthyridine 75, a potent and selective tau aggregate binder with potential as a tau PET tracer.

Simple Ligand-Receptor Interaction Descriptor (SILIRID) for alignment-free binding site comparison.

We describe SILIRID (Simple Ligand-Receptor Interaction Descriptor), a novel fixed size descriptor characterizing protein-ligand interactions. SILIRID can be obtained from the binary interaction fingerprints (IFPs) by summing up the bits corresponding to identical amino acids. This results in a vector of 168 integer numbers corresponding to the product of the number of entries (20 amino acids and one cofactor) and 8 interaction types per amino acid (hydrophobic, aromatic face to face, aromatic edge to face, H-bond donated by the protein, H-bond donated by the ligand, ionic bond with protein cation and protein anion, and interaction with metal ion). Efficiency of SILIRID to distinguish different protein binding sites has been examined in similarity search in sc-PDB database, a druggable portion of the Protein Data Bank, using various protein-ligand complexes as queries. The performance of retrieval of structurally and evolutionary related classes of proteins was comparable to that of state-of-the-art approaches (ROC AUC ≈ 0.91). SILIRID can efficiently be used to visualize chemogenomic space covered by sc-PDB using Generative Topographic Mapping (GTM): sc-PDB SILIRID data form clusters corresponding to different protein types.

Molecular modeling of ligand-receptor interactions in GABA C receptor.

A new homology model of the GABA binding site of the GABA(C) receptor was built. Natural agonist GABA and antagonist TPMPA were docked into the receptor and molecular dynamics simulation of the complexes was performed to clarify binding poses of the ligands. It was shown that orientation of the ligand is defined by salt bridges between the ligand and the arginine (Arg104) and glutamate residues (Glu194 and Glu196) of the binding site. Different behavior and binding poses for agonist and antagonist was demonstrated by molecular dynamics simulation along with differential movement of the loop C during agonist and antagonist binding. Binding orientations of the ligands revealed that main binding forces in the GABA binding site should be electrostatic ones.