Determination of Geographical Origin of Agricultural Products by Smart Metabolites Database™

Introduction

Mislabeling of the geographical origin of agricultural products is a universal problem in the agricultural and food products industries, and various techniques for determination of the geographical origin have been developed to date. Since determination of geographical origin is generally difficult based on genetic information and other bioinformation possessed innately by agricultural products, information on acquired substances in agricultural products is used. In this type of determination, techniques based on information of elements introduced in agricultural products from external sources have excellent stability and have been used in research for many years. In addition to substances introduced from external sources, the acquired substances in agricultural products also include substances that are produced internally, such as amino acids, organic acids, fatty acids, and sugars. These substances, called metabolites as a general term, are contained in common in many agricultural products, and their concentration is thought to change dynamically at a timing determined by the surrounding environment, even in the same type of agricultural product. Therefore, if it is possible to discover patterns in the abundance ratio of the metabolites in a designated type of agricultural product in various geographic regions by comprehensive measurement of the metabolites in that agricultural product, it should be possible to use those patterns to determine the geographical origin of products produced in different regions. In this article, the metabolites in a total of 106 samples of domestic Japanese asparagus and asparagus produced in other countries were measured using Shimadzu Smart Metabolites Database, which enables simultaneous measurement of the compositions of 337 hydrophilic metabolites, and a model for determination of domestic or foreign origin was prepared. The results of this experiment confirmed that determination with accuracy of approximately 90% is possible.

December 4, 2019 GMT