Multivariate correlation of infrared fingerprints and molecular weight distributions with bioactivity of poultry by-product protein hydrolysates

Characterization of protein hydrolysates is a vital step in developing peptide-based bioactive ingredients. Multivariate correlation of chemical fingerprints and bioactivity of poultry by-product protein hydrolysates is explored as a potential analytical strategy for characterization and quality control. Chemical fingerprints of sixty hydrolysates were acquired using Fourier-transform infrared spectroscopy (FTIR) and size exclusion chromatography (SEC). Bioactivities (2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging and angiotensin-1-converting enzyme (ACE-1) inhibition) were measured in vitro. Partial least squares regression models based on FTIR fingerprints or SEC chromatograms showed a better prediction performance for ACE-1 inhibition (coefficients of determination (R2) = 0.91, root mean square error of prediction (RMSECV) = 2.8; R2 = 0.85, RMSECV = 3.5, respectively) than for DPPH radical scavenging (R2 = 0.74, RMSECV = 0.3; R2 = 0.75, RMSECV = 0.3, respectively). Such models are promising tools for rapid prediction of bioactivities and as a quality control technology in production of bioactive peptides.