Comparing Metabolites and Functional Properties of Various Tomatoes Using Mass Spectrometry-Based Metabolomics Approach

Abstract

Tomato is one of the world's most consumed vegetables, and thus, various cultivars have been developed. Therefore, metabolic differences and nutrient contents of various tomatoes need to be discovered. To do so, we performed metabolite profiling along with evaluation of morphological and physicochemical properties of five representative tomato types. Common tomato cultivars, bigger and heavier than other tomatoes, contained higher levels of amino acids, organic acids, and lipids. On the contrary, cherry tomato cultivars contained a higher proportion of phenylpropanoids, lycopene, β-carotene, and α-carotene than the other tomatoes. Also, the highest antioxidant activity and total phenolic and flavonoid contents were observed in cherry tomato cultivars. Furthermore, to understand metabolic distributions in various tomato cultivars, we constructed a metabolic pathway map. The higher metabolic flux distribution of most primary metabolite synthetic pathways was observed in common tomatoes, while cherry tomato cultivars showed a significantly elevated flux in secondary metabolite synthetic pathways. Accordingly, these results provide valuable information of different characteristics in various tomatoes, which can be considered while purchasing and improving tomato cultivars.

Keywords: MS-based metabolomics approach; functional properties; metabolic pathway analysis; physicochemical properties; tomato.

Figure 1

Photographs of five types of tomato.

Figure 2

Results of antioxidant activities [2,2′-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) (A) and ferric reducing antioxidant power (FRAP) (B)] in five types of tomato. Different letters in the bar graph indicate significant difference by ANOVA followed by Duncan's multiple range test (p-value < 0.05). TEAC, trolox equivalent antioxidant capacity; CT, chal tomato; ST, tomato; KT, kumato; CH, cherry tomato; JT, jujube-shaped cherry tomato.

Figure 3

Principal component analysis (PCA) (A,B) score plots for metabolites in five types of tomato based on GC-TOF-MS (A) and UHPLC-LTQ-Orbitrap-MS/MS (B) data set. Heat map analysis of five tomato types based on GC-TOF-MS (C) and UHPLC-LTQ-Orbitrap-ESI-MS/MS (D) data. Heat map representation of the relative abundance of significantly discriminant metabolites (VIP > 0.7, p-value < 0.05) based on PLS-DA model (Supplementary Figure 1). Chal tomato (), tomato (), kumato (), cherry tomato (), and jujube-shaped cherry tomato () samples. GC-TOF-MS, gas chromatography time-of-flight mass spectrometry; UHPLC-LTQ-Orbitrap-MS/MS, ultrahigh-performance liquid chromatography–linear trap quadrupole-orbitrap–tandem mass spectrometry; VIP, variable importance projection; PLS-DA, partial least squares-discriminant analysis.

Figure 4

Schematic diagram of the metabolic pathway and relative levels of metabolites in five types of tomato. The relative levels are shown as fold-changes normalized using the average of all values. The metabolic pathway was modified based on the KEGG database (http://www.genome.jp/kegg/). C, chal tomato; T, tomato; K, kumato; CH, cherry tomato; JT, jujube-shaped cherry tomato.

Figure 5

Correlation map between the metabolite levels and observed ABTS, FRAP, weight, and width in five types of tomato. Each metabolite is identified as significantly different metabolites through PLS-DA model (Supplementary Figure 1). Each square implies Pearson's correlation coefficient between metabolites and assayed activities. The red color indicates a positive (0 < r < 1) correlation, and the blue colors indicate a negative (−1 < r < 0) correlation. Asterisks indicate p-values < 0.05. ABTS, 2,2′-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt; FRAP, ferric reducing antioxidant power; PLS-DA, partial least squares-discriminant analysis.