Multi-Layer Computational Design of Small-Molecule Inhibitors Targeting Nck1 for Autoimmune Disease Treatment

Abstract

The adaptor protein Nck1 involved in the T cell CD3 signaling pathway is a promising therapeutic target for the treatment of immune system-related diseases. This study established a multi-layer virtual screening pipeline that integrates compound library generation, molecular docking, pan-assay inference compounds (PAINS) filtration, and free energy calculation from molecular dynamics (MD) simulation to design and sequentially identify promising novel small molecule inhibitors (SMIs) for autoimmune diseases treatment. The compound library consists of 10,149 potential SMIs by performing R group enumeration on a first-in-class SMI of Nck1, AX-024, proven effective in empirical experimentation. The compounds were first screened using molecular docking followed by PAINS analysis to narrow down the design pool for higher level computational modeling and simulations. The remaining compounds were subjected to MD simulations for binding free energy calculation of the SMI-Nck1 systems, which was used to propose top candidates with stable binding. For these candidates, extended MD trajectories were produced to reveal structural mechanics behind their strong performance. Binding mechanism analysis, RMSF profiling, and evaluation of R-group effects identified promising molecules, such as d6294, that consistently outperformed the reference AX-024 in simulations. These analyses revealed key interactions with the Nck1 protein, including salt bridges, cation-π interactions, and hydrogen bonds, that underpin their stability and potential effectiveness. Taken together, this study establishes a transferable pipeline for the in silico design of SMIs and identifies several top candidates with strong potential for subsequent in vitro validation for the treatment of Nck1-related immune system diseases.