Proline-Based Structural Rules for Predicting Prolyl Endopeptidase Inhibitory Peptides from Food Proteins: In Vitro Validation and Multiscale Computational Characterization

Prolyl endopeptidase (PEP) is a therapeutic target for neurodegenerative disorders, yet systematic approaches for discovering PEP inhibitory peptides from food proteins remain limited. This study developed a four-rule proline-based screening framework─evaluating proline presence (A1), hydrophobic–proline motifs (A2), consecutive proline patterns (A3), and overlap with known inhibitors (A4)─and applied it to 243 food proteins across five enzyme treatments via in silico hydrolysis. Fourteen candidate peptides were synthesized and validated in vitro (IC50 = 2.12–68.50 μM), with A1 showing strong correlation with inhibitory potency (r = 0.831, p < 0.001). Molecular dynamics, Molecular Mechanics Poisson–Boltzmann Surface Area (MM-PBSA), and quantum mechanical analyses revealed a peptide-length confounding effect on binding energy and demonstrated that potency depends on concentrated, high-quality interactions at the catalytic site rather than total contact extent. This framework provides a practical prioritization strategy for identifying PEP inhibitory peptides from food protein hydrolysates.