International Journal of Food Properties (Dec 2022)

Polyphenols, flavonoids, and antioxidant content of honey coupled with chemometric method: geographical origin classification from Amhara region, Ethiopia

  • Marie Yayinie,
  • Minaleshewa Atlabachew,
  • Alemu Tesfaye,
  • Woldegiorgis Hilluf,
  • Chaltu Reta,
  • Tessera Alemneh

DOI
https://doi.org/10.1080/10942912.2021.2021940
Journal volume & issue
Vol. 25, no. 1
pp. 76 – 92

Abstract

Read online

The Amhara region of Ethiopia is endowed with honey of diverse varieties and qualities. However, there is a lack of information on its secondary metabolite content and antioxidant nature. For this study, 47 fresh honey samples were collected from seven administrative zones throughout three provinces in the Amhara National Regional State, Ethiopia. The honey samples’ restorative nature was evaluated by the quantitative determination of phenolic content and antioxidant capabilities using standard colorimetric methods. The finding showed that the mean values of total polyphenol content based on gallic acid equivalent (GAE) ranged from 17.03 to 42.04 mg GAE per 100 g of honey. The mean value of the entire flavonoid content using catechin equivalent (CE) was from 3.20 to 7.40 mg CE, and when using quercetin equivalent (QE), it ranged from 1.67 to 5.08 mg QE, per 100 g of honey sample. The ascorbic acid equivalent antioxidant content (AEAC) of the honey samples – ranged from 16.23 to 26.5 9 mg AEAC per 100 g of the honey samples. The percent antioxidant activities (% AA) of the honey samples—ranged from 23.74 to 40.11%. There was a strong positive correlation between phenolic content and antioxidant properties. Amber-colored honey enjoyed the highest value on the stated parameters based on the samples’ colors, while the white-colored samples registered the least value. Based on the findings, it is concluded that the region’s honey has a magnificent therapeutic nature. Using the principal component analysis (PCA) model, the top three principal components described 96.63% of the total variations. The linear discriminant analysis (LDA) model has an average of 68.1% discriminant power. The LDA model was cross-validated by the leave-one-out cross-validation approach, and 70.21% of it clustered adequately. In the biplot analysis, honey sample distribution based on their color clustered better than the geographic origin and climate factors.

Keywords