Enhancing Drug Discovery and Development through the Integration of Medicinal Chemistry, Chemical Biology, and Academia-Industry Partnerships: Insights from Roche's Endocannabinoid System Projects.

The endocannabinoid system (ECS) is a critical regulatory network composed of endogenous cannabinoids (eCBs), their synthesizing and degrading enzymes, and associated receptors. It is integral to maintaining homeostasis and orchestrating key functions within the central nervous and immune systems. Given its therapeutic significance, we have launched a series of drug discovery endeavors aimed at ECS targets, including peroxisome proliferator-activated receptors (PPARs), cannabinoid receptors types 1 (CB1R) and 2 (CB2R), and monoacylglycerol lipase (MAGL), addressing a wide array of medical needs. The pursuit of new therapeutic agents has been enhanced by the creation of specialized labeled chemical probes, which aid in target localization, mechanistic studies, assay development, and the establishment of biomarkers for target engagement. By fusing medicinal chemistry with chemical biology in a comprehensive, translational end-to-end drug discovery strategy, we have expedited the development of novel therapeutics. Additionally, this strategy promises to foster highly productive partnerships between industry and academia, as will be illustrated through various examples.

Δ-Quantum machine-learning for medicinal chemistry.

Many molecular design tasks benefit from fast and accurate calculations of quantum-mechanical (QM) properties. However, the computational cost of QM methods applied to drug-like molecules currently renders large-scale applications of quantum chemistry challenging. Aiming to mitigate this problem, we developed DelFTa, an open-source toolbox for the prediction of electronic properties of drug-like molecules at the density functional (DFT) level of theory, using Δ-machine-learning. Δ-Learning corrects the prediction error (Δ) of a fast but inaccurate property calculation. DelFTa employs state-of-the-art three-dimensional message-passing neural networks trained on a large dataset of QM properties. It provides access to a wide array of quantum observables on the molecular, atomic and bond levels by predicting approximations to DFT values from a low-cost semiempirical baseline. Δ-Learning outperformed its direct-learning counterpart for most of the considered QM endpoints. The results suggest that predictions for non-covalent intra- and intermolecular interactions can be extrapolated to larger biomolecular systems. The software is fully open-sourced and features documented command-line and Python APIs.

Development of High-Specificity Fluorescent Probes to Enable Cannabinoid Type 2 Receptor Studies in Living Cells.

Pharmacological modulation of cannabinoid type 2 receptor (CB2R) holds promise for the treatment of numerous conditions, including inflammatory diseases, autoimmune disorders, pain, and cancer. Despite the significance of this receptor, researchers lack reliable tools to address questions concerning the expression and complex mechanism of CB2R signaling, especially in cell-type and tissue-dependent contexts. Herein, we report for the first time a versatile ligand platform for the modular design of a collection of highly specific CB2R fluorescent probes, used successfully across applications, species, and cell types. These include flow cytometry of endogenously expressing cells, real-time confocal microscopy of mouse splenocytes and human macrophages, as well as FRET-based kinetic and equilibrium binding assays. High CB2R specificity was demonstrated by competition experiments in living cells expressing CB2R at native levels. The probes were effectively applied to FACS analysis of microglial cells derived from a mouse model relevant to Alzheimer's disease.

NMR pseudocontact shifts in a symmetric protein homotrimer.

NMR pseudocontact shifts are a valuable tool for structural and functional studies of proteins. Protein multimers mediate key functional roles in biology, but methods for their study by pseudocontact shifts are so far not available. Paramagnetic tags attached to identical subunits in multimeric proteins cause a combined pseudocontact shift that cannot be described by the standard single-point model. Here, we report pseudocontact shifts generated simultaneously by three paramagnetic Tm-M7PyThiazole-DOTA tags to the trimeric molecular chaperone Skp and provide an approach for the analysis of this and related symmetric systems. The pseudocontact shifts were described by a "three-point" model, in which positions and parameters of the three paramagnetic tags were fitted. A good correlation between experimental data and predicted values was found, validating the approach. The study establishes that pseudocontact shifts can readily be applied to multimeric proteins, offering new perspectives for studies of large protein complexes by paramagnetic NMR spectroscopy.

Mechanical Stabilization of Helical Chirality in a Macrocyclic Oligothiophene.

We introduce a design principle to stabilize helically chiral structures from an achiral tetrasubstituted [2.2]paracyclophane by integrating it into a macrocycle. The [2.2]paracyclophane introduces a three-dimensional perturbation into a nearly planar macrocyclic oligothiophene. The resulting helical structure is stabilized by two bulky substituents installed on the [2.2]paracyclophane unit. The increased enantiomerization barrier enabled the separation of both enantiomers. The synthesis of the target helical macrocycle 1 involves a sequence of halogenation and cross-coupling steps and a high-dilution strategy to close the macrocycle. Substituents tuning the energy of the enantiomerization process can be introduced in the last steps of the synthesis. The chiral target compound 1 was fully characterized by NMR spectroscopy and mass spectrometry. The absolute configurations of the isolated enantiomers were assigned by comparing the data of circular dichroism spectroscopy with TD-DFT calculations. The enantiomerization dynamics was studied by dynamic HPLC and variable-temperature 2D exchange spectroscopy and supported by quantum-chemical calculations.

Exploring protein-ligand binding affinity prediction with electron density-based geometric deep learning.

Rational structure-based drug design relies on accurate predictions of protein-ligand binding affinity from structural molecular information. Although deep learning-based methods for predicting binding affinity have shown promise in computational drug design, certain approaches have faced criticism for their potential to inadequately capture the fundamental physical interactions between ligands and their macromolecular targets or for being susceptible to dataset biases. Herein, we propose to include bond-critical points based on the electron density of a protein-ligand complex as a fundamental physical representation of protein-ligand interactions. Employing a geometric deep learning model, we explore the usefulness of these bond-critical points to predict absolute binding affinities of protein-ligand complexes, benchmark model performance against existing methods, and provide a critical analysis of this new approach. The models achieved root-mean-squared errors of 1.4-1.8 log units on the PDBbind dataset, and 1.0-1.7 log units on the PDE10A dataset, not indicating significant advantages over benchmark methods, and thus rendering the utility of electron density for deep learning models context-dependent. The relationship between intermolecular electron density and corresponding binding affinity was analyzed, and Pearson correlation coefficients r > 0.7 were obtained for several macromolecular targets.

Geometric deep learning-guided Suzuki reaction conditions assessment for applications in medicinal chemistry.

Suzuki cross-coupling reactions are considered a valuable tool for constructing carbon-carbon bonds in small molecule drug discovery. However, the synthesis of chemical matter often represents a time-consuming and labour-intensive bottleneck. We demonstrate how machine learning methods trained on high-throughput experimentation (HTE) data can be leveraged to enable fast reaction condition selection for novel coupling partners. We show that the trained models support chemists in determining suitable catalyst-solvent-base combinations for individual transformations including an evaluation of the need for HTE screening. We introduce an algorithm for designing 96-well plates optimized towards reaction yields and discuss the model performance of zero- and few-shot machine learning. The best-performing machine learning model achieved a three-category classification accuracy of 76.3% (±0.2%) and an F 1-score for a binary classification of 79.1% (±0.9%). Validation on eight reactions revealed a receiver operating characteristic (ROC) curve (AUC) value of 0.82 (±0.07) for few-shot machine learning. On the other hand, zero-shot machine learning models achieved a mean ROC-AUC value of 0.63 (±0.16). This study positively advocates the application of few-shot machine learning-guided reaction condition selection for HTE campaigns in medicinal chemistry and highlights practical applications as well as challenges associated with zero-shot machine learning.

G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction.

Protein-ligand interactions (PLIs) determine the efficacy and safety profiles of small molecule drugs. Existing methods rely on either structural information or resource-intensive computations to predict PLI, casting doubt on whether it is possible to perform structure-free PLI predictions at low computational cost. Here we show that a light-weight graph neural network (GNN), trained with quantitative PLIs of a small number of proteins and ligands, is able to predict the strength of unseen PLIs. The model has no direct access to structural information about the protein-ligand complexes. Instead, the predictive power is provided by encoding the entire chemical and proteomic space in a single heterogeneous graph, encapsulating primary protein sequence, gene expression, the protein-protein interaction network, and structural similarities between ligands. This novel approach performs competitively with, or better than, structure-aware models. Our results suggest that existing PLI prediction methods may be improved by incorporating representation learning techniques that embed biological and chemical knowledge.

Enabling late-stage drug diversification by high-throughput experimentation with geometric deep learning.

Late-stage functionalization is an economical approach to optimize the properties of drug candidates. However, the chemical complexity of drug molecules often makes late-stage diversification challenging. To address this problem, a late-stage functionalization platform based on geometric deep learning and high-throughput reaction screening was developed. Considering borylation as a critical step in late-stage functionalization, the computational model predicted reaction yields for diverse reaction conditions with a mean absolute error margin of 4-5%, while the reactivity of novel reactions with known and unknown substrates was classified with a balanced accuracy of 92% and 67%, respectively. The regioselectivity of the major products was accurately captured with a classifier F-score of 67%. When applied to 23 diverse commercial drug molecules, the platform successfully identified numerous opportunities for structural diversification. The influence of steric and electronic information on model performance was quantified, and a comprehensive simple user-friendly reaction format was introduced that proved to be a key enabler for seamlessly integrating deep learning and high-throughput experimentation for late-stage functionalization.

Prospective de novo drug design with deep interactome learning.

De novo drug design aims to generate molecules from scratch that possess specific chemical and pharmacological properties. We present a computational approach utilizing interactome-based deep learning for ligand- and structure-based generation of drug-like molecules. This method capitalizes on the unique strengths of both graph neural networks and chemical language models, offering an alternative to the need for application-specific reinforcement, transfer, or few-shot learning. It enables the "zero-shot" construction of compound libraries tailored to possess specific bioactivity, synthesizability, and structural novelty. In order to proactively evaluate the deep interactome learning framework for protein structure-based drug design, potential new ligands targeting the binding site of the human peroxisome proliferator-activated receptor (PPAR) subtype gamma are generated. The top-ranking designs are chemically synthesized and computationally, biophysically, and biochemically characterized. Potent PPAR partial agonists are identified, demonstrating favorable activity and the desired selectivity profiles for both nuclear receptors and off-target interactions. Crystal structure determination of the ligand-receptor complex confirms the anticipated binding mode. This successful outcome positively advocates interactome-based de novo design for application in bioorganic and medicinal chemistry, enabling the creation of innovative bioactive molecules.