Metabolite profile is a powerful analytical technique to identify the functional characterization of plants. In this study, the phytochemicals and secondary metabolites of lentils (
Legumes are high-quality and low-cost foods that can replace animal proteins (Mira
The three kind of lentils used in this study were commercial varieties which were imported from America and had the biggest market share. These were French green whole lentil (FG), red whole lentil (LR), and green whole lentil (LG) (Fig. 1). These samples were purchased from ASIA SEED Co., LTD. Fresh samples were freeze-dried for at least 72 hours and then crushed to a fine powder with the Planetary Mono Mill (Pulverisette 6; Fritsch, Idar-Oberstein, Germany). The powder was stored at −80°C before analysis.
Polar metabolites were extracted from 0.1 g samples using 1 mL of 2.5 : 1 : 1 v/v/v methanol/water/chloroform. Ribitol (0.06 mL, 0.2 mg/mL of water) was added as an internal standard (IS). The extraction was performed at 37°C with a mixing frequency of 1,200 rpm for 30 minutes using a Thermomixer (compact model 5355; Eppendorf AG, Hamburg, Germany). The solutions were centrifuged at 16,000 ×
The fatty acid composition of the lipophilic fraction was analyzed according to the method described by AOCS (1997). The fatty acid content was determined using lipid extraction and saponification with 0.5 N sodium hydroxide in methanol. The saponification mixture was methylated with 14% boron trifluoride/methanol and the resulting methyl esters were extracted with pentane. Methyl esters of the fatty acids were analysed by GC (GC 7890B; Agilent, Atlanta, GA, USA). Pentadecanoic acid (1 mg/mL) was used as an IS.
Carotenoids were extracted from 0.1 g lentil seed samples by adding 3 mL of ethanol containing 0.1% ascorbic acid (w/v), vortex mixing, and incubation in a water bath at 85°C for 5 minutes. The carotenoid extract was saponified with potassium hydroxide (0.12 mL, 80% w/v) in a water bath at 85°C for 10 minutes. After saponification, the samples were immediately placed on ice, and cold deionized water (1.5 mL) was added. β-Apo-8′-carotenal (0.2 mL, 25 μg/mL) was added as an IS. To separate the layers, carotenoids were extracted twice with hexane (1.5 mL) by centrifugation at 1,200 ×
Flavonoids were extracted from 0.1 g of lentil seed powder by adding 1.2 mL of 50% methanol containing 1.2 M HCl in a water bath at 80°C for 2 hours. The crude suspensions were centrifuged at 10,000 ×
Anthocyanin extraction was performed according to a slightly modified method of Kim
Soluble (free and esterified forms) and insoluble (bound form) phenolic acids were extracted according to the procedure described by Mira
The powdered samples (0.01 g) were mixed with 1 mL of methanol and incubated at 30°C for 30 minutes at a mixing frequency of 1,200 rpm using a thermomixer comfort. After incubation, the supernatants were centrifuged at 10,000 rpm at 4°C for 10 minutes. and then, 1 mL of methanol was added. This procedure was repeated. An extract dilution series of 25, 50, and 100% was made in methanol and pure methanol (1 mL) was used as a control. To compare the levels of scavenging activity, a dilution series of 2,6-di-
All determinations were repeated five times for reproducibility. Experimental data were analysed by the Duncan multiple-range test (SAS 9.2; SAS Institute, Cary, NC, USA) and analysis of variance (ANOVA). Principal component analysis (PCA) models for each components were developed by using for soft independent modeling of class analogy (BioPAT-SIMCA, ver. 13_Umetrics, Umeå, Sweden). Technically, a principal component was defined as a linear combination of optimally-weighted observed variables. In the course of performing a principal component analysis, it is possible to calculate a score for each subject on a given principal component. The PCA output consisted of score plots to visualize the contrast between different samples. The data file was scaled using unit variance scaling before all variables were subjected to PCA. Pearson correlation analysis was conducted with the SAS 9.2 software package (SAS Institute). Correlation analysis was conducted on the relative levels of 73 metabolites with standardization pre-processing. HCA and heat-map visualization of the correlation coefficient were performed using MultiExperiment Viewer version 4.4.0 software (http://www.tm4.org/mev/).
Hydrophilics with low molecular weight such as amino acids, organic acids, sugars, and sugar alcohols were analyzed by using GC-TOF MS. 42 peaks of untargeted metabolomics were identified through comparisons with related compounds in the library database of the Chroma TOF software (Fig. 2a). Relationships between metabolites were identified through Pearson correlation analysis and HCA. Metabolites were quantified based on the peak area ratio to an IS. The internal standard signal intensities obtained from the 42 metabolites were normalized based on PCA analysis to determine the variation in metabolite profiles between the cultivars. PCA models from data of polar metabolites of the three lentil varieties were to assess whether the lentil varieties showed separation according to the different color with different metabolite profiles. PCA score plots showed a distinct separation by different colored variety and the total deviation value of components 1 (61.6%) and 2 (28.7%) was 90.3% (Fig. 2b). The identification and profiling of primary metabolites using GC-TOFMS analysis allows clear discrimination between lentil genotypes. The loading plot of 42 components (20 amino acids, 7 sugars, 4 sugar alcohols and 11 organic acids) was shown in Fig. 2c. PCA scores of shikimic acid of aromatic amino acid and trehalose of disaccharide were close to 0 showing independence. In addition, six amino acids (4-aminobutyric acid, proline, serine, asparagine, glutamine, and arginine) showed positive PCA score values, whereas three organic acids (glyceric acid, nicotinic acid, and succinic acid) and four amino acids (threonic acid, tryptophan, glutamic acid, and pyroglumic acid) had negative values. Even though they were present in small amounts, shikimic acid and trehalose were identified only in FG and LG, respectively, whereas mannose was detected only in FG and LG. From these results, metabolite profiles could be affected to the color characterization showing clear separation in the PCA chemometrics.
Compared to other legumes, lentils are low in fat; therefore, it has been used as a plant protein source in weight-loss diets (Dueñas
All carotenoids including flavonoids, and anthocyanins were analysed by HPLC, on the basis of which the metabolic differences among the three genotypes and the relationships with their colors were analyzed. Carotenoids, which are fat-soluble pigments, mainly confer yellow, orange, and red colors to plants and play an important role in human nutrition as precursors of vitamin A and anti-oxidants, of which lutein and zeaxanthin are important for eye health and protect cells from carcinogens (Wang
Flavonoids are vacuolar water-soluble pigments that are classified into flavonols, flavones, flavanones, catechins, anthocyanins, and chalcones (Kim
To estimate the anti-oxidative capacity of the three lentil varieties, contents of six phenolic acids (
DPPH radical scavenging activity of extracts was measured to investigate the antioxidant activity of the three lentil varieties. BHT, a synthetic oxidizer, was used as a control. Based on measurements of the optical density of the reaction mixture, the EC50, which is the concentration of sample required to inhibit 50% of DPPH radicals, was calculated, with lower EC50 indicating higher antioxidant activity (Table 2). The EC50 values were within the range of 2.78–6.32 mg/mL DW. DPPH activities of FG and LG were 2.9 mg/mL DW and 2.78 mg/mL DW, respectively, and were stronger than that of LR. The DPPH assay confirmed that lentil extracts with higher phenolic and flavonoid contents had stronger antiradical action (Alshikh
Correlation analysis and the HCA were performed to examine the relationships among metabolites analyzed from different colored lentils (Fig. 4). The result of HCA could illustrated the extent of distances among compositional nutrients or among lentil varieties, respectively. Applied variables were based on their mutual correlation coefficients from SAS analysis. Distances between samples are typically displayed in a dendrogram which provides a simple graphical view of sample groupings; the length of the branches (cluster distance) is a measure of the degree of similarity between metabolites or groups of samples. The longer distance showed in the dendrogram meant the lower correlation between analyzed variables (Fig. 4). In the horizontal line listed with analyzed metabolites, the correlation value (r) between asparagine and glucose was 0.10617, which were lowest in the result of correlation analysis. This means that these two variables are less influenced by each other. The correlations between myristic acid and succinic acid (r = 0.49927) and between inositol and isolecine (r = 0.41493) were also low, in order, showing relatively longer distances in the result of HCA dendrogram (Fig. 4). In the vertical dendrogram, the distances between FG and LG, between LG and LR, and between FG and LR were longer showing lower correlation values than those between variables in the same varieties showing higher correlation values. It could be demonstrated that the lentils with different colors by different color varieties. These results may also offer opportunities for further omics studies related to the safety assessment of new biotechnology plants.
This study was aimed to compare the metabolite profiles related to antioxidant components in the various lentils with different colors by investigating primary and derivative secondary metabolites. Among 42 hydrophilic low molecular weight compounds, shikimic acid and trehalose were found only in the FG and LG variety, respectively. Mannose was found exclusively in the FG and LG varieties. In the fatty acid contents, LR variety showed highest contents, which was contrary result to the phenolics and DPPH activity. Three kind of lentils were not different statistically with each other in the lutein (
This study was supported by the National Institute of Agricultural Science (Code PJ01175201), Rural Development Administration, Republic of Korea. We are also thankful to the reviewers whose comments led to substantial improvements in this paper and to the Executive Editor for his consideration.
Download Form