The development of selective small-molecule inhibitors against extracellular signal-regulated kinases 1 and 2 (ERK1/2) remains a major challenge due to their high structural similarity and conformational flexibility. In this study, we combined classical molecular dynamics (MD) simulations with artificial intelligence (AI)–based unsupervised learning approaches to investigate the conformational dynamics and binding selectivity of ERK1/2-inhibitor complexes. Six systems, comprising ERK1 and ERK2 bound to inhibitors 33A, 38Z, and Z48, were simulated for 250 ns to capture their structural and energetic behaviors. Root mean square deviation (RMSD) and fluctuation (RMSF) analyses confirmed that all complexes reached equilibrium, with ERK2 systems exhibiting greater structural stability than ERK1. Radius of gyration (Rg) and hydrogen-bond profiles support consistent compactness and persistent interactions throughout the trajectories. The AI-based dimensionality reduction (principal component analysis PCA, t-distributed stochastic neighbor embedding t-SNE) and K-means clustering revealed distinct conformational basins, indicating ligand-specific modulation of ERK flexibility. Free energy landscape mapping and state transition analyses demonstrated that Z48-bound systems showed more compact and stable conformational basins, reflecting enhanced conformational stabilization, whereas 33A complexes exhibited higher dynamic variability. In contrast, molecular mechanics/generalized Born surface area (MM/GBSA) binding free energy calculations showed that 38Z exhibits the strongest binding affinity toward both ERK isoforms, driven primarily by favorable van der Waals interactions. Overall, this combined MD-AI framework provides atomistic insights into ERK1/2 conformational plasticity and inhibitor selectivity, highlighting 38Z as the most energetically favorable binder and Z48 as the most effective conformational stabilizer. These findings advance the understanding of ERK regulation and can guide the rational design of next-generation selective kinase inhibitors.
Lama A. Alqahtani (Sun,) studied this question.