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.

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.

Monoacylglycerol lipase (MAGL) regulates endocannabinoid 2-arachidonoylglycerol (2-AG) and eicosanoid signalling. MAGL inhibition provides therapeutic opportunities but clinical potential is limited by central nervous system (CNS)-mediated side effects. Here, we report the discovery of LEI-515, a peripherally restricted, reversible MAGL inhibitor, using high throughput screening and a medicinal chemistry programme. LEI-515 increased 2-AG levels in peripheral organs, but not mouse brain. LEI-515 attenuated liver necrosis, oxidative stress and inflammation in a CCl₄-induced acute liver injury model. LEI-515 suppressed chemotherapy-induced neuropathic nociception in mice without inducing cardinal signs of CB₁ activation. Antinociceptive efficacy of LEI-515 was blocked by CB₂, but not CB₁, antagonists. The CB₁ antagonist rimonabant precipitated signs of physical dependence in mice treated chronically with a global MAGL inhibitor (JZL184), and an orthosteric cannabinoid agonist (WIN55,212-2), but not with LEI-515. Our data support targeting peripheral MAGL as a promising therapeutic strategy for developing safe and effective anti-inflammatory and analgesic agents.

Despite its essential role in the (patho)physiology of several diseases, CB₂R tissue expression profiles and signaling mechanisms are not yet fully understood. We report the development of a highly potent, fluorescent CB₂R agonist probe employing structure-based reverse design. It commences with a highly potent, preclinically validated ligand, which is conjugated to a silicon-rhodamine fluorophore, enabling cell permeability. The probe is the first to preserve interspecies affinity and selectivity for both mouse and human CB₂R. Extensive cross-validation (FACS, TR-FRET and confocal microscopy) set the stage for CB₂R detection in endogenously expressing living cells along with zebrafish larvae. Together, these findings will benefit clinical translatability of CB₂R based drugs.

(1) Background: The cannabinoid 2 receptor (CB₂R) is a promising anti-inflammatory drug target and development of selective CB₂R ligands may be useful for treating sight-threatening ocular inflammation. (2) Methods: This study examined the pharmacology of three novel chemically-diverse selective CB₂R ligands: CB₂R agonists, RO6871304, and RO6871085, as well as a CB₂R inverse agonist, RO6851228. In silico molecular modelling and in vitro cell-based receptor assays were used to verify CB₂R interactions, binding, cell signaling (ß-arrestin and cAMP) and early absorption, distribution, metabolism, excretion, and toxicology (ADMET) profiling of these receptor ligands. All ligands were evaluated for their efficacy to modulate leukocyte-neutrophil activity, in comparison to the reported CB₂R ligand, HU910, using an in vivo mouse model of endotoxin-induced uveitis (EIU) in wild-type (WT) and CB₂R^-/- mice. The actions of RO6871304 on neutrophil migration and adhesion were examined in vitro using isolated neutrophils from WT and CB₂R^-/- mice, and in vivo in WT mice with EIU using adoptive transfer of WT and CB₂R^-/- neutrophils, respectively. (3) Results: Molecular docking studies indicated that RO6871304 and RO6871085 bind to the orthosteric site of CB₂R. Binding studies and cell signaling assays for RO6871304 and RO6871085 confirmed high-affinity binding to CB₂R and selectivity for CB₂R > CB₁R, with both ligands acting as full agonists in cAMP and ß-arrestin assays (EC₅₀s in low nM range). When tested in EIU, topical application of RO6871304 and RO6871085 decreased leukocyte-endothelial adhesion and this effect was antagonized by the inverse agonist, RO6851228. The CB₂R agonist, RO6871304, decreased in vitro neutrophil migration of WT neutrophils but not neutrophils from CB₂R^-/-, and attenuated adhesion of adoptively-transferred leukocytes in EIU. (4) Conclusions: These unique ligands are potent and selective for CB₂R and have good immunomodulating actions in the eye. RO6871304 and RO6871085, as well as HU910, decreased leukocyte adhesion in EIU through inhibition of resident ocular immune cells. The data generated with these three structurally-diverse and highly-selective CB₂R agonists support selective targeting of CB₂R for treating ocular inflammatory diseases.