Geographical Origin Differentiation of Rice by LC–MS-Based Non-Targeted Metabolomics

Abstract

Many factors, such as soil, climate, and water source in the planting area, can affect rice taste and quality. Adulterated rice is common in the market, which seriously damages the production and sales of high-quality rice. Traceability analysis of rice has become one of the important research fields of food safety management. In this study, LC–MS-based non-targeted metabolomics technology was used to trace four rice samples from Heilongjiang and Jiangsu Provinces, namely, Daohuaxiang (DH), Huaidao No. 5 (HD), Songjing (SJ), and Changlixiang (CL). Results showed that the discrimination accuracy of the partial least squares discriminant analysis (PLS-DA) model was as high as 100% with satisfactory prediction ability. A total of 328 differential metabolites were screened, indicating significant differences in rice metabolites from different origins. Pathway enrichment analysis was carried out on the four rice samples based on the KEGG database to determine the three metabolic pathways with the highest enrichment degree. The main biochemical metabolic pathways and signal transduction pathways involved in differential metabolites in rice were obtained. This study provides theoretical support for the geographical origins of rice and elucidates the change mechanism of rice metabolic pathways, which can shed light on improving rice quality control.

Geographical origin traceability of rice using a FTIR-based metabolomics approach

“…After this, VIP 4 1.5 principle continues to screen 'markers'. Previous metabolomics studies 36,39 rank the metabolites as 'biomarkers' in order of VIP values, if we also adopted this principle, eligible 'markers' would probably be on show by a series of scattered and erratic wave number values, it is inconvenient to compare the 'marker' differences among various rice groups. To better cope with this issue, we chose the longest sequence of 'markers' ranked in consecutive number of variables as the characteristic band to distinguish the specific rice group from others.…”

“…57 In this study, there are 1623 variables and 153 samples, conforming to the characteristics of far more variables than samples in metabolomics analysis. [35][36][37][38][39][40][47][48][49] The OPLS-DA classification model to deal with metabolomics data may incur over-fitting, 58 that is to say, the model is able to distinguish samples well in the training set, but underperforms to predict the new sample sets. Hence, it is essential to verify the reliability of the OPLS-DA model.…”

“…A great advantage of the metabolomics approach is that it can achieve accurate traceability when the number of samples is much smaller than the variables. [35][36][37][38][39][40] Theoretically speaking, it is feasible to use the metabolomics method to achieve the origin traceability by finding the metabolites (i.e., 'biomarkers') of an organism with significant differences between a control group and an experimental group, that is to say, just two group samples are enough to realize the origin traceability of an organism. [35][36][37][38][39][40] To date, metabolomics has widely been used in the origin traceability of rice.…”

“…[35][36][37][38][39][40] Theoretically speaking, it is feasible to use the metabolomics method to achieve the origin traceability by finding the metabolites (i.e., 'biomarkers') of an organism with significant differences between a control group and an experimental group, that is to say, just two group samples are enough to realize the origin traceability of an organism. [35][36][37][38][39][40] To date, metabolomics has widely been used in the origin traceability of rice. [35][36][37][38][39][40] In this study, we try to find wave numbers with absorbance significantly different between the specific rice group and other rice groups as 'wave number markers', which enlighten us from Fourier-Transform InfraRed (FTIR) spectroscopy recognized as a crucial tool for metabolic fingerprinting, [41][42][43][44] as it is capable of analyzing proteins, polysaccharides, carbohydrates, fatty acids, lipids and nucleic acids rapidly and simultaneously with a minimum amount of sample preparation or background training.…”

“…[35][36][37][38][39][40] To date, metabolomics has widely been used in the origin traceability of rice. [35][36][37][38][39][40] In this study, we try to find wave numbers with absorbance significantly different between the specific rice group and other rice groups as 'wave number markers', which enlighten us from Fourier-Transform InfraRed (FTIR) spectroscopy recognized as a crucial tool for metabolic fingerprinting, [41][42][43][44] as it is capable of analyzing proteins, polysaccharides, carbohydrates, fatty acids, lipids and nucleic acids rapidly and simultaneously with a minimum amount of sample preparation or background training.…”