Discrimination of Geographical Origin of Agricultural Products From Small-Scale Districts by Widely Targeted Metabolomics With a Case Study on Pinggu Peach

Abstract

Geographical indications of agricultural products are characterized by high quality and regional attributes, while they are more likely to be counterfeited by similar products from nearby regions. Accurate discrimination of origin on small geographical scales is extremely important for geographical indications of agricultural products to avoid food fraud. In this study, a widely targeted metabolomics based on ultra-high-performance liquid chromatography-tandem mass spectrometry combined with multivariate statistical analysis was used to distinguish the geographical origin of Pinggu Peach of Beijing and its two surrounding areas in Heibei province (China). Orthogonal partial least squares-discriminant analysis (OPLS-DA) based on 159 identified metabolites showed significant separation from Pinggu and the other adjacent regions. The number of the most important discriminant variables (VIP value >1) was up to 62, which contributed to the differentiation model. The results demonstrated that the metabolic fingerprinting combined with OPLS-DA could be successfully implemented to differentiate the geographical origin of peach from small-scale origins, thus providing technical support to further ensure the authenticity of geographical indication products. The greenness of the developed method was assessed using the Analytical GREEnness Metric Approach and Software (ARGEE) tool. It was a relatively green analytical method with room for improvement.

Keywords: Analytical GREEnness metric approach; metabolic fingerprint; origin discrimination; peach; small-scale districts; widely targeted metabolomics.

Figure 1

Geographical map of peach sampling sites from different regions in China and the specific localities of peach samples within Pinggu (Beijing, capital city), Shunping (Hebei province), and Laoting (Hebei province).

Figure 2

PCA score plot of test samples and quality control (QC) samples (, QC sample; test sample).

Figure 3

Orthogonal partial least squares-discriminant analysis (OPLS-DA) test for the identification of the three regions of peach samples. PG (Pinggu, Beijing ), SP (Shunping, Hebei ), and LT (Laoting, Hebei ).

Figure 4

Permutation test of OPLS-DA model for Pinggu group vs. Shunping group.

Figure 5

Volcano plot and hierarchical cluster analysis (HCA), and Venn diagram. (A,B) represent the volcano plot showing the differential expression of metabolites between PG and SP, and PG and LT, respectively. The red dots in the figure represent the differentially expressed metabolites that were increased, the blue dots represent the decreased differentially expressed metabolites, and the gray color indicates the differentially expressed metabolites that were not significant. (C,D) represent the heat map showing the differential expression of metabolites between PG and SP, and PG and LT, respectively. The red color indicates the increase in differentially expressed metabolites, while the blue color indicates the decrease in differentially expressed metabolites. (E) is the Venn diagram showing the specific and common differential metabolites between PG and SP, and PG and LT.

Figure 6

KEGG annotations and enrichment of differentially expressed metabolites of each pairwise comparison of peach. (A) PG vs. SP; (B) PG vs. LT. Bubbles represent metabolic pathways, and the size and horizontal coordinate of the bubble represent the influence factor in the pathway. The larger the bubble size, the greater the impact factor. The longitudinal coordinate and the color of the bubble represent the P-value of enrichment analysis. The deeper the color and the smaller the P-value represent more significant enrichment degree.

Figure 7

Greenness of the developed method was assessed with Analytical GREEnness metric approach (AGREE); annotated result of a generic assessment (above) and the corresponding color scale for reference (below).