Facilitated identification of bioactive peptide fractions and optimization of enzymatic protein hydrolysis using size-exclusion chromatography fingerprints: Combining interval PLS and response surface modeling

Protein hydrolysates made from food industry byproducts have the potential to be used as nutraceuticals that can alleviate high blood pressure, by inhibiting angiotensin converting enzyme I (ACE-I). Discovery and optimization of bioactive peptide fractions, in typically complex protein hydrolysates, is a laborious and time-consuming task. In the present study, a library of 108 hydrolysates were produced from mechanical deboning chicken residue, using three different temperatures, four protease concentrations and nine hydrolysis times, in a full factorial manner. Employed design of experiment enabled a thorough screening of the experimental space and resulted in hydrolysates with substantial variation in bioactivity. Hydrolysates were characterized by size-exclusion chromatography and by in vitro ACE-I inhibition assay. To identify a bioactive fraction, interval-partial least square (iPLS) regression based on a chromatographic fingerprint of a crude samples was used, which eliminated the need for time-consuming bioassay-guided fractionation. After the bioactive fraction was identified, the response surface modeling (RSM) was employed to develop a predictive model for the optimization of hydrolysate properties by adjusting the production parameters. Here we are presenting a novel methodology combining: (1) SEC fingerprint-based iPLS for facilitated identification of bioactive peptide fractions, and (2) RSM using the identified SEC fractions as a response for a robust optimization of the hydrolysis process. A combination of iPLS and RSM utilizing chromatographic fingerprints was demonstrated as a promising analytical approach that can facilitate discovery of bioactive peptides in complex hydrolysates and further enable optimization of production conditions.