Computational Discovery and Optimization of a Potent, Selective, and Drug-Like Scaffold for p38α Inhibition.

Purpose: The aim of this study was to discover and optimize a novel chemical scaffold capable of selectively inhibiting p38α, a kinase involved in inflammatory and neurodegenerative diseases. Despite decades of work, most p38α inhibitors have failed clinically due to limited selectivity, compensatory signaling, and safety issues. We sought to combine computational and experimental approaches to identify potent, drug-like, and selective inhibitors suitable for further development. Methods: A consensus virtual screening workflow (ESSENCE-Dock), integrating DiffDock, LeadFinder, and GNINA, was applied to the Eurofins-Villapharma compound library. The top hit guided similarity searching and clustering to identify related analogues for structure-activity relationship studies. Binding modes and substituent contributions were analyzed using molecular modeling and molecular dynamics simulations. Biochemical HTRF assays, ADME profiling, NanoBRET intracellular target engagement, and kinome-wide screening were used to evaluate potency, cellular activity, and selectivity. Results: Virtual screening identified a previously unreported 3,5-disubstituted dihydropyrazolo[1,5- Conclusion: This study identifies and characterizes a novel and drug-like p38α inhibitor scaffold with potent biochemical activity, high kinome selectivity, and confirmed intracellular target engagement. The combined computational-experimental workflow provides a strong foundation for further optimization toward therapeutic candidates for inflammatory and neurodegenerative diseases.