NMR Metabolite Profiling in the Quality and Authentication Assessment of Greek Honey-Exploitation of STOCSY for Markers Identification

Abstract

Honey is a natural, healthy commodity and is probably among the most complex foods produced by nature. It is the oldest recorded and certainly the only natural sweetener that can be used by humans without any further processing. Nowadays, the increase in honey's value, along with its growing list of healthy attributes, has made the present raw material a prime target for adulteration. In the current study, NMR-based metabolite profiling in combination with chemometrics was applied in the quality control of Greek honeys from northeastern Aegean islands. Moreover, statistical total correlation spectroscopy (STOCSY) was employed for the first time as a dereplication and structural elucidation tool in the honey biomarker identification process. A total of 10 compounds were successfully identified in honey total extracts via ¹H NMR spectroscopy. Compounds such as 5-(hydroxymethyl)furfural, methyl syringate, a mono-substituted glycerol derivative and 3-hydroxy-4-phenyl-2-butanone, among others, were identified as potential biomarkers related to the botanical and geographical origin of the samples. High-Resolution Mass Spectrometry (HRMS) was used as an additional verification tool on the identified compounds.

Keywords: Greek honey; NMR profiling; STOCSY; biomarkers; botanical origin; geographical origin.

Figure 1

Representative 1D ¹H NMR spectra from samples of different botanical origins: (a) thyme, (b) plant, (c) blossom and (d) heather honey. Assigned metabolites: 5-hydroxymethylfurfural (5-HMF), 1; methyl syringate (MSYR), 2; 3-hydroxy-4-phenyl-2-butanone, 3; monosubstituted glycerol derivative (MG), 4; unedone, 5; 1-(4-methoxyphenyl)-ethane-1,2-diol, 6; 4-methoxybenzoic acid (4-MBA), 7; benzoic acid (BZA), 8; abscisic acid (ABA), 9; dehydrovomifoliol, 10.

Figure 2

PLS-DA scores scatter plot of the dataset showing clustering based on (a) botanical and (b) geographical origin.

Figure 3

Statistical total correlation spectroscopy (STOCSY) 1D-pseudo-NMR spectra. (a) 5-HMF, “driver peak” at 4.663 ppm; (b) MSYR, “driver peak” at 7.263 ppm; (c) 3-hydroxy-4-phenyl-2-butanone, “driver peak” at 7.229 ppm; and (d) MG, “driver peak” at 4.133 ppm.

Figure 4

Box plots of statistically significant biomarkers based on their botanical origin are presented. (a) 5-HMF, (b) 3-hydroxy-4-phenyl-2-butanone, (c) MSYR, (d) ABA, (e) 1-(4-methoxyphenyl)-ethane-1,2-diol and (f) 4-methoxybenzoic acid. All p-values can be found in Table S6.

Figure 5

Box plots of statistically significant biomarkers based on their geographical origin (p-value ≤ 0.05 was considered as statistically significant). All p-values can be found in Tables S7 and S8.