Identification of compounds from chufa (Eleocharis dulcis) peels by widely targeted metabolomics

Abstract The Chinese water chestnut (CWC) is among the most widespread and economically important vegetables in Southern China. There are two different types of cultivars for this vegetable, namely, big CWC (BCWC) and small CWC (SCWC). These are used for different purposes based on their metabolic profiles. This study aimed to investigate the metabolite profile of CWC and compare the profiles of peels collected in different harvest years using ultraperformance liquid chromatography/mass spectrometry (UPLC–MS)‐based metabolomics analysis. Three hundred and twenty‐one metabolites were identified, of which 87 flavonoids, 25 phenylpropanoids, and 33 organic acids and derivatives were significantly different in the content of the two varieties of BCWC and SCWC. The metabolite profiles of the two different cultivars were distinguished using principle component analysis (PCA) and orthogonal projections to latent structures discriminant analysis, and the results indicated differences in the metabolite profile of Eleocharis dulcis (Burm. f.) Trin. ex Hensch. Three isomers of hydroxycoumarin, namely, O‐feruloyl‐4‐hydroxycoumarin, O‐feruloyl‐3‐hydroxycoumarin, and O‐feruloyl‐2‐hydroxycoumarin, exhibited increased levels in BCWC, while p‐coumaric acid and vanillic acid did not show any significant differences in their content in BCWC and SCWC peels. This study, for the first time, provides novel insights into the differences among metabolite profiles between BCWC and SCWC.

and inhibit acrylamide formation and are often used in traditional Chinese medicine to treat pharyngitis, laryngitis, enteritis, cough, hepatitis, and hypertension (Luo et al., 2014;Nie et al., 2018). BCWC is one of the most popular hydrophytic vegetables in China, owing to its unique flavor . Fresh-cut Chinese water chestnut has been used to preserve food products and beverages and is sold worldwide Luo et al., 2014).
The SCWC is rich in starch and is usually not consumed directly.
Currently, it is used for vermicelli production during food processing (Tang et al., 2018), and leads to the production of large amounts of SCWC peels as waste. The peel of CWC is rich in bioactive components, which is benefit to human body. However, there are still few studies on the chemical constituents of CWC peels (Luo et al., 2014).
At present, there are a few reports regarding the types of flavonoids and phenolic acids present in the peels of BCWC and SCWC.
Qualitative and quantitative transformation of nutrients and bioactive compounds is difficult. Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) is a rapid and highly sensitive method (Raclariu et al., 2017;Wang et al., 2019) that is used for the detection of metabolic products in comparison databases (Barraza-Elenes et al., 2019;Frolov et al., 2013;Fang et al., 2003) and was used to identify and quantify metabolites present in BCWC and SCWC peels, especially flavonoids and organic acids and their derivatives.
In this study, we conducted an ultraperformance liquid chromatography/mass spectrometry (UPLC/MS)-based metabolomics analysis to investigate the constituents of two CWC cultivars. The chemical compositions of the two cultivars of CWC were distinguished here, which is of great significance to the future utilization of CWC peels.

| Plant materials and samples
BCWC and SCWC were planted in the same field and under the same conditions at the same time, which is the typical size of the selected Chinese varieties (Fanglin Chinese water chestnut, harvested from Guangxi Province in southern China in November 2018), and identified by Prof. Yanghe Luo, Hezhou University (Hezhou, Guangxi, China). The peels of BCWC and SCWC were stored at −80°C until further analysis.

| Sample preparation and extraction
Freeze-dried BCWC and SCWC peels were crushed using a mixer mill (MM 400, Retsch, Germany) with a zirconia bead for 1.5 min at 30 Hz. Hundred milligrams of the powder was weighed and extracted overnight at 4°C using 1.0 ml of 70% aqueous methanol.

| ESI-Q TRAP-MS/MS analysis
Linear ion trap (LIT) and triple quadrupole (QQQ) scans were performed using a QQQ TRAP API Q TRAP LC/MS/MS System, equipped with an ESI Turbo Ion-Spray interface, operating in the positive ion mode and controlled by Analyst 1.6.3 software (AB Sciex). Operation parameters of the ESI source were as follows: ion source, turbo spray; source temperature 500°C; ion spray voltage (IS) 5500 V; ion source gas I (GSI), gas II (GSII), and curtain gas (CUR) were set at 55, 60, and 25.0 psi, respectively; the collision gas (CAD) was high. Instrument tuning and mass calibration were performed with 10 and 100 μmol/L polypropylene glycol solutions in QQQ and LIT modes, respectively. QQQ scans were acquired in the multiple reaction monitoring (MRM) mode with the collision gas (nitrogen) set to 5 psi. DP and CE for individual MRM transitions were determined and further optimized (Chen et al., 2013). A specific set of MRM transitions was monitored for each period in accordance with metabolites eluted within said period.

| Qualitative and quantitative analysis of metabolites
Qualitative analyses were conducted using the stepwise multiple-ion monitoring-enhanced product ions (MIM-EPI) strategy and the MS2T data were analyzed by comparing the accurate precursor ion (Q1) and product ion (Q3) values, retention time (RT), and a self-compiled database MWDB (Met Ware biological science and Technology Co., Ltd) and publicly available metabolite databases if the standards were unavailable (Chen et al., 2013;Wei et al., 2014). Isotope signals, repeated signals containing K + , Na + , and NH + , and repeated signals of further fractionated substances, can be determined through qualitative analysis of substances in accordance with their spectral data of K + , Na + , and NH + obtained during fractionation. The results show that the fractionation of the fractionated substances is the same as that of the other substances in the fractionation (Pinnapat et al., 2018).

| Multivariate and cluster analysis of BCWC and SCWC peels
All 321 metabolomics were analyzed. To eliminate the effect of concentration on pattern recognition, the logarithm (log10) of the peak area matrix of BCWC and SCWC metabolites was determined, which was followed by Poisson normalization (Xia et al., 2015). Thereafter, cluster analysis of the metabolite profiles of BCWC and SCWC peels was conducted using LC-MS/MS analysis. Results obtained using BCWC and SCWC peels were dichotomized as follows (Figure 1a): values for BCWC and SCWC peels were segregated in the PCA score plot of sesame metabolites. Furthermore, they were clearly divided into two classes on the heat map, indicating significant differences in levels of secondary metabolites in BCWC and SCWC peels.

| Differential metabolite analysis based on PCA
Principal component analysis (PCA) is often used to study the internal structure of multiple variables using a few principal components or to derive a few principal components from the original variable so that they can retain as much information about the original variable as possible. Moreover, these components/variables are not related to each other, and usually, a mathematical formula is used to represent the original indicators as a linear combination or a new comprehensive index. In the PCA plot (Figure 4a), PC1 and PC2 were 57.1% and 15.46%, respectively. ED 1 and ED 2 showed a clear distinction between the samples, indicating that there is a large difference in metabolites between BCWC and SCWC.

| Differential metabolite analysis via partial least squares-discriminant analysis (OPLS-DA)
OPLS-DA is more sensitive to variables with low correlations (Thévenot et al., 2015). The constituents of SCWC and BCWC were compared to identify the metabolites responsible for the observed F I G U R E 2 Phenylpropanoids content in BCWC and SCWC differences. OPLS-DA models were used to carry out pairwise comparisons of these metabolites. High predictability (Q 2 ) and strong goodness of fit (R 2 X, R 2 Y) of the OPLS-DA models were observed on comparing BCWC and SCWC (Q 2 = 0.963, R 2 X = 0.599, R 2 Y = 0.995,

| Differential metabolic pathways in BCWC and SCWC
Differential metabolites between BCWC and SCWC were mapped to the KEGG database to obtain detailed information regarding metabolic pathways they may participate in ( Figure S2)

| CON CLUS ION
In the present study, the chemical profiles of BCWC and SCWC were analyzed using the widely targeted metabolomics method and metabolite compounds were identified and classified.
Marked differences were observed in the metabolites of SCWC and BCWC. In total, 321 differential metabolites were identified, comprising flavonoids, phenylpropanoids, organic acids, and F I G U R E 4 PCA and OPLS-DA of the relative differences in metabolites in different CWC cultivars. a: Score plots for principle components 1 and 2 showed high cohesion within groups and good separation between two CWC cultivars from the BCWC and SCWC. The sampling groups are color coded as follows: Green = ED1; orange = ED2; and mustard = QC samples. b: 0PLS-DA model plots and loading plots for the BCWC (ED1) and SCWC (ED2). c: The volcano plot shows the differential metabolite expression levels between BCWC and SCWC. Green dots represent downregulated differentially expressed metabolites; red spots represent upregulated differentially expressed metabolites, and gray represents insignificantly expressed metabolites derivatives. In particular, 15 alkaloids and 13 terpenes were identified, for the first time, in Eleocharis dulcis (Burm. f.) Trin. ex Hensch.
CWC is among the most widespread and economically important vegetables, with a wide array of uses in the food and medicine industries. To the best of our knowledge, the difference in chemical profiles between different cultivars may inform different functions. Our work focused on the different flavonoids and phenylpropanoids contained in the two different CWC cultivars. In total, 87 flavonoids were identified with different levels in BCWC and SCWC. We also identified 25 phenylpropanoids, of which higher levels were found in SCWC than in BCWC. Feruloyl and dihydroxycoumarin were the predominant phenylpropanoids in SCWC. Our results help further the current understanding of metabolic mechanisms accounting for the differences in different cultivars of Eleocharis dulcis (Burm. f.) Trin. ex Hensch.

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data will be available from the authors upon request.

E TH I C S S TATEM ENT
This research does not include any human or creature testing.