Comparative Metabolomics study of flavonoids in the pericarp of different coloured bitter gourds (Momordica charantia L.)

Abstract

Bitter gourd (Momordica charantia L.) is a member of Cucurbitaceae family and has long been used as a source of food and medicine for its rich bioactive components or secondary metabolites. However, there are relatively few large-scale detection, identification, and quantitative studies on flavonoids in the pericarp of bitter gourds of different colours. To determine the differences in the diversity and specificity of flavonoids in the pericarp of bitter gourd of different colours, the metabolic profiles in the pericarp of three coloured bitter gourd accessions, dark green (mo), pale green (lv), and white (bai), were analysed by ultra-performance liquid chromatography-tandem mass spectrometry. Priorly, it was confirmed that the different shades of green were caused by the content of chlorophyll. A total of 93 metabolites, including 90 flavonoids and three tannins, were detected in the current study. These 90 flavonoids included three isoflavones, nine dihydroflavones, seven flavanols, 34 flavonols, 26 flavonoids, four chalcones, five flavonoid carbonosides, and two dihydroflavonols. Compared to mo, both lv and bai had 21 and 25 different metabolites, respectively, while there were only nine different metabolites between lv and bai. The relative contents of vitexin and isovitexin increased with the deeper colour of the bitter gourd. Thus, the different metabolites in coloured bitter gourds are mainly involved in the biosynthesis of flavonols, flavonoid carbonosides, and flavonoids. This study enables identification of metabolic differences in the pericarp of bitter gourds of different colours. The results will be helpful for quality breeding of new bitter gourd varieties and shall provide a reference for their medical application.

Supplementary information: The online version contains supplementary material available at 10.1007/s12298-022-01210-7.

Keywords: Bitter gourd; Chlorophyll content; Metabolome; Pericarp.

Fig. 1.

Fruits harvested at 18 days after simultaneous pollination (A) and total chlorophyll content (B) in the pericarp of three bitter gourd accessions; mo: black green; Iv: green; bai:white. TheY axis corresponds to the chlorophyll content of three biological repetitions. ****: P ≤ 0.0001, ***: P ≤ 0.001

Fig. 2

Differences in the relative contents of the tannin in fruits of the three bitter gourd accessions 18 days after simultaneous pollination. The Y axis correspond to the average of metabolite data log10-transformed. ***: P ≤ 0.001

Fig. 3

Diffeernces in the relative contents of the dihydroflavonol, dihydroflavone, flavonold carbonoside in fruits of the three bitter gourd accessions 18 days after simultaneous pollination. The Y axis corresponds to the average of metabolite data log10-transformed. *:P ≤ 0.05, **: P ≤ 0.01, ***: P ≤ 0.001, ***:P ≤ 0.0001

Fig. 4

Differences in the relatives content of flavonols, flavonoid in fruits of the three bitter gourd accessions 18 days after simultaneous pollination. The Y axis corresponds to the average of metabolite data log10-transformed. ****:P ≤ 0.0001. **: P ≤ 0.01. *: P ≤ 0.05

Fig. 5

Heat map and PCA of the relative differences in the metabolites profiles among three bitter gourd with different color. A Heat map visualization. The content of each metabolite was normalized to complete linkage hierarchical clustering. Each sample is represent by one column, and each flavonoid is visualized in one row. Red indicates high abundance, whereas low relative metabolites are shown in green (color key scale right of the heat map). B Score plots of principal components 1 and 2 demonstrated high cohesion within groups and good separation among three bitter gourd accessions mix, QC samples, the mixture of the bitter gourd sample extracts

Fig. 6

OPLS-DA model plots and loading plots of pairwise comparisons between the three different color bitter gourd

Fig. 7

The volcano plot shows the differential metabolite expression levels between lv, bai and mo respectively. Green dots represent downregulated differentially expressed metabolites: red spots represented upregulated differentially expressed metabolites; and black represented non-differentially expressed metabolites

Fig. 8

Venn diagram showing the overlapping and accessions pecific differential metabolites

Fig. 9

Enrichment of the differential metabolites to distinct KEGG pathways. Differential metabolites in lv, mo compared with those in bai were mapped to distinct metabolic pathways. Enriched P-values were computed from a hypergeometric distribution. AP-value cutoff of 0.01 was selected to reduce the FDR