Nontargeted Metabolomics as a Screening Tool for Estimating Bioactive Metabolites in the Extracts of 50 Indigenous Korean Plants

Abstract

Many indigenous Korean plants have been used in medicinal preparations and health-promoting foods. These plant species contain beneficial metabolites with various bioactivities, such as antioxidant and anti-inflammatory activities. Herein, we suggest a new screening strategy using metabolomics to explore the bioactive compounds in 50 Korean plants. Secondary metabolites were analyzed using UHPLC-LTQ-Orbitrap-MS/MS. The plant extracts were subjected to antioxidant and anti-inflammatory assays. We identified metabolites that contributed to bioactivities according to the results of bioassays and multivariate analyses. Using Pearson's correlation, phenolics (e.g., casuarictin, 3-O-methylellagic acid) showed positive correlation with antioxidant activity, while biflavonoids (e.g., amentoflavone, rosbustaflavone) were correlated with nitric oxide (NO) inhibition activity. To compensate for the limitation of this new strategy, we further validated these by investigating three parts (branches, fruits, leaves) of Platycladus orientalis which showed high activities on both bioassays. Unlike the above observation, we identified significantly different metabolites from different parts, which was not the results of bioassays. In these validation steps, interestingly, biflavonoids (e.g., robustaflavone, sciadopitysin) contributed to both activities in P. orientalis. The findings of this work suggest that new strategy could be more beneficial in the identification of bioactive plant species as well as that of their corresponding bioactive compounds that impart the bioactivity.

Keywords: UHPLC-LTQ-Orbitrap-MS/MS; anti-inflammatory activity; antioxidant activity; indigenous plant; metabolite profiling.

Figure 1

(a) PCA score plot; (b) PLS biplot derived from the UHPLC-LTQ-Orbitrap-MS dataset for Table 1. indigenous plants. PLS biplot shows the correlation between the metabolite variations and the selected test bioactivities in the extracts of the 50 indigenous plants. The samples were analyzed using three analytical replicates for each sample. ▲: Plant species which have high antioxidant activities; ▲: Plant species which have high NO inhibition activity; ▲: Plant species which include in 50 indigenous plants.

Figure 2

Permutation plots of the PLS model describing the R2 and Q2 Y-intercepts for NO inhibitory (a), ABTS (b), and DPPH (c) activities of 50 indigenous plant extracts. The PLS model was validated using 200 permutation tests to evaluate its goodness of fit and predictive power.

Figure 3

Heatmap analysis in plant species which have high NO inhibitory activities derived from UHPLC-LTQ-Orbitrap-MS/MS data. The heatmap indicates the relative contents in the secondary metabolites among the different fractions. ^a Metabolites that have a high contribution to the bioactivities were determined by PLS-biplot (VIP > 1.0, p < 0.05). N.I.: Non-identified metabolite.

Figure 4

Heatmap analysis in plant species which have high antioxidant activities derived from UHPLC-LTQ-Orbitrap-MS/MS data. The heatmap indicates the relative contents in the secondary metabolites among the different fractions. ^a Metabolites that have high contribution to the bioactivities were determined by PLS-biplot (VIP > 1.0, p < 0.05). N.I.: Non-identified metabolite.

Figure 5

Correlation map between the relative abundance of the secondary metabolites and (a) NO production inhibitory activity and (b) antioxidant (ABTS, DPPH) activity of the extracts of 50 indigenous plants. Each square indicates Pearson’s correlation coefficient values (r). Red and blue represent positive (0 < r < 0.5) and negative (−0.5 < r < 0) correlations, respectively. * FDR-adjusted p-value < 0.05; ^a Metabolites that have a high contribution to the bioactivities were determined by PLS-biplot (VIP > 1.0). N.I.: Non-identified metabolites.

Figure 6

(a) PCA and (b) PLS-DA score plots of the results obtained for the different parts from P. orientalis analyzed by UHPLC-LTQ-Orbitrap-MS/MS. The samples were analyzed using three analytical replicates for each sample. ●: Fruits of P. orientalis, ●: Leaves of P. orientalis, ●: Branches of P. orientalis.

Figure 7

Results of the (a) NO inhibitory assay, (b) MTT assay, and the antioxidant activity assays, namely (c) ABTS and (d) DPPH of the extracts of the different parts from P. orientalis. Values of (a,b) are expressed as the mean ± standard deviation (SD) of four biological replicates. Values of (c,d) are expressed with three biological replicates. Bar graph denoted by the same letter were not significantly different according to Duncan’s multiple range test (p < 0.05). Control of (a,b) indicates LPS alone without treating plant extracts.

Figure 8

Correlation map between the relative abundance of the significantly different metabolites and the results of the antioxidant assay (ABTS and DPPH), and NO inhibition assay of the extracts of the different parts of P. orientalis. Each square indicates Pearson’s coefficient values (r). Red and blue represent positive (0 < r < 1.0) and negative (−1.0 < r < 0) correlations, respectively. *: p-value < 0.05.