AI-assisted identification and in vitro validation of a MyD88 pathway modulator for inflammatory airway signalling.

Pediatric obstructive sleep apnea-hypopnea syndrome (OSAHS) associated with allergic rhinitis is characterized by persistent airway inflammation, in which the TLR4-MyD88-NF-κB signaling pathway plays a central role. Targeting this pathway may provide an effective strategy for modulating inflammatory responses. In the present study, an integrated computational and experimental approach was employed to identify and validate potential modulators of MyD88 signaling. De novo ligand design was performed to generate initial candidates, followed by artificial intelligence (AI)-based fragmentation and optimization to improve binding affinity and drug-like properties. Molecular docking and molecular dynamics simulations were conducted to evaluate binding stability and interaction profiles. The most promising compound was synthesized and structurally characterized using standard analytical techniques. Biological evaluation was performed across multiple in vitro models, including epithelial (RPMI 2650), macrophage (RAW 264.7), and mast cell (HMC-1) systems. Functional assays demonstrated that the optimized compound maintained cell viability while significantly reducing intracellular reactive oxygen species (ROS) levels. Western blot and quantitative PCR analyses revealed downregulation of key signaling components, including TLR4, MyD88, IRAK4, TRAF6, and NF-κB. In addition, ELISA-based cytokine profiling showed reduced levels of pro-inflammatory (TNF-α, IL-6, IL-1β) and Th2 cytokines (IL-4, IL-5, IL-13). Collectively, the findings indicate that the AI-fragmented lead compound was associated with attenuation of the TLR4-MyD88-NF-κB signaling axis across multiple biological levels. While further in vivo validation is required, this study highlights the potential of integrating AI-assisted drug design with experimental validation for the development of targeted therapeutics in inflammatory airway disorders.