Non-destructive analysis of anthocyanin content and antioxidant activity in single purple corn kernel by near-infrared spectroscopy

Abstract

The objective of this study was to explore the potential of near-infrared reflectance (NIR) spectroscopy to determine single kernel composition in purple corn. NIR spectra and analytical measurements of anthocyanin contents and antioxidant activity (1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging activity) were collected from 120 single purple corn kernels. Partial least-squares (PLS) regression models were developed with 90 purple corn accessions randomly assigned to calibration data set and 30 accessions randomly assigned to an external validation set. PLS regression for anthocyanin contents and DPPH radical-scavenging activity had sufficient accuracy for kernel sorting applications, with the external validation set having a standard error of prediction (SEP) =3.04 μmol Cy3Glc eq/g D.W. and 1.66 μmol Trolox eq/g D.W. The validation correlation and standard deviation/square error of validation values (RPD) and the coefficient of determination for validation (R ²_v) were determined to be as follows: anthocyanin contents, 0.84 and 2.5; DPPH radical-scavenging activity, 0.87 and 2.7. These present results indicated that prediction of anthocyanin content and DPPH radical-scavenging activity of single purple corn kernel could be measured using NIR spectroscopy, nondestructively.