The corticotropin-releasing factor (CRF) and its type 1 receptor (CRF₁R) play a key role in the regulation of the hypothalamic–pituitary–adrenal (HPA) axis. Dysregulation of the HPA axis is associated with congenital adrenal hyperplasia (CAH) and depression. Non-peptide CRF₁R-selective antagonists displayed antidepressant effects on animal models and are used for the management of CAH. To develop novel non-peptide CRF₁R antagonists, we have previously designed and synthesized a series of substituted pyrimidines. Among these analogs, molecule 43 (M43) binds to CRF₁R with the highest affinity. Based on this finding, we selected M43 for further pharmacological characterization in the present study. The results suggest that M43 is a potent CRF₁R antagonist, blocking the ability of the CRF-related agonist, Tyr⁰-sauvagine, to stimulate (1) cAMP accumulation in HEK 293 cells expressing CRF₁R and (2) the proliferation rate of RAW 264.7 macrophages. Computational studies suggest that the antagonist properties of M43 are mostly attributed to its ability to interact with residues in the allosteric pocket of CRF₁R, comprised of the third, fifth, and sixth transmembrane domain residues, which block activation-associated structural rearrangements of the receptor. Our data will be used to design novel non-peptide CRF₁R antagonists for clinical use.

Drug repurposing is a state-of-the-art pipeline that expands pharmaceutical research and development portfolia. Despite computational tools are rapidly advancing, they predominantly rely on the examination of textual (1D) data. Herein, we present Cloudscreen ^®, a "one-stop-shop" platform for drug repurposing, which seamlessly integrates 3D (structural data through predictive protein-ligand modeling) and 1D data alongside state-of-the-art machine learning algorithms. Our platform leverages a knowledge graph database, curated from diverse sources, including publicly available repositories and databases coupled to in-house computations. We are harnessing the power of artificial intelligence to predict and assess the efficacy and safety of repurposed biomolecules for novel therapeutic indications while interrogating the human variome. Moreover, Cloudscreen ^® expands such prediction capabilities based on AlphaFold models, ADMETox and pharmacogenomics (emphasis on missense variants). Cloudscreen ^® is a powerful tool that results from the synergy of wet- and dry-lab validated datasets and hence, provides “one-stop-shop” predictive insights into the uncharted therapeutic possibilities for repurposed drugs.