Tailoring the physicochemical diversity of plant-derived peptide pools through protease selection and hydrolysis strategy

Abstract

Proteolytic hydrolysis is a primary route for generating peptide mixtures from plant proteins, yet the influence of protease selection on the physicochemical composition of the resulting hydrolysates remains systematically under characterised. This study evaluated seven commercially available proteases: Trypsin, Alcalase, Streptomyces griseus protease, Elastase, Molsin, Glu-C, and Proteinase K, across single-enzyme, concurrent dual-enzyme, and sequential dual-enzyme hydrolysis setups, applied to structurally complex rhizome-derived plant protein matrices (Zingiber officinale, Daucus carota, Raphanus sativus, Ipomoea batatas). A total of 10,049 unique peptides were identified across 15 hydrolysis conditions by LC-MS/MS, of which 9,198 were condition-specific. Peptide identification counts differed significantly across all conditions (Kruskal–Wallis H = 119.74, p = 7.07 × 10⁻¹⁹), with Alcalase generating the highest mean count (674 ± 290.7 peptides per replicate), representing a 6.3-fold increase over Trypsin (107.6 ± 67.4 peptides per replicate). Beyond identification count, protease selection significantly and consistently modulated key physicochemical attributes of the resulting peptide pools. GRAVY (Grand Average of Hydropathicity) score distributions differed significantly across conditions (H = 149.74, p < 0.0001), with sequential strategies incorporating Trypsin producing the most hydrophobic peptide pools relative to Trypsin alone (p_BH < 0.001). Hydrophobic residue content was significantly elevated across all multi-enzyme strategies compared to Trypsin (p_BH < 0.05), while sequential combinations consistently led to identification of longer peptides (mean 13.9–15.5 aa vs. 10.7 aa for Trypsin; p_BH < 0.001). Terminal residue analysis confirmed that cleavage specificity directly determines compositional patterns at peptide termini, with enzyme-specific features clearly distinguishable across all conditions. These findings demonstrate that protease selection is a tunable upstream variable with statistically significant effects on the physicochemical landscape of plant protein hydrolysates, providing an evidence-based framework for designing hydrolysis workflows targeted toward specific downstream applications in food ingredient development or functional peptide screening.