From predicted interfaces to measured affinity: mapping a growth-factor receptor binding mechanism relevant to...

To delineate the molecular recognition features of the FGF7-FGFR2b signaling axis relevant to meniscus regeneration, we integrated AI-based structural modeling (AlphaFold3), mutational energetics, molecular dynamics (MD) simulations, and isothermal titration calorimetry (ITC). Structural analysis of the predicted FGF7-FGFR2b complex revealed a stable interfacial hydrogen-bond network. In silico alanine scanning identified three critical receptor hotspot residues-Lys25, Arg112, and Asp144-whose substitutions were predicted to significantly destabilize the interface by elevating mutation energies (> 1.5 kcal/mol) and reducing polar-contact retention. MD simulations confirmed that both single and triple hotspot mutations (K25A/R112A/D144A) led to increased conformational fluctuations and impaired complex stability relative to the wild-type. Quantitative binding assays via ITC validated these findings, demonstrating a high-affinity interaction for the wild-type complex ($K = 2.31 \times 10^7$ L/mol), whereas each hotspot mutation markedly reduced the binding constant. These results map a discrete receptor recognition patch essential for FGF7 coordination and provide a structural foundation for modulating the STAT/MAPK signaling pathway to enhance fibrocartilage repair during meniscus healing.