The complexity of chemicals in Traditional Chinese Medicine (TCM) including
Network pharmacology is an impressive methodology for investigating the systematic pharmacological mechanism through the constructing and analyzing biological networks such as protein-protein interaction, chemical-target-pathway network, which could provide direction for the further discovery of new drug without enormous time, money, and effort [
Therefore, this current study was designed to develop a fast and effective method for the chemical characterization and systematic pharmacological mechanism of
The chemical reference standards isoquercitrin, luteolin, naringenin, and kaempferol were provided by Chengdu Herbpurify biotechnology CO., LTD (Chengdu, China); Phlorizin, Phloretin, and Trilobatin were purchased from Sichuan Wei Keqi Biotechnology Co., Ltd (Sichuan, China); eriodictyol and hyperoside were provided by Chengdu Push biotechnology CO., LTD (Chengdu, China); quercetin and apigenin were provided by Chengdu Alfa biotechnology CO., LTD (Chengdu, China); baicalein and wogonin were obtained from Chengdu Desite biotechnology CO., LTD (Chengdu, China). The purities of all chemical reference standards were above 98% by HPLC-DAD.
Acetonitrile and methanol of chromatography grade were provided by MERCK (Darmstadt, Germany); The ultrapure water was produced by a milli-Q water purification system (Millipore, Milford, MA, United States); formic acid of LC-MS grade and all other reagents of analytical grade were purchased from Aladdin Industrial Corporation.
A total of 10 g powdered root of
The reference standards including hyperoside, isoquercitrin, phlorizin, eriodictyol, trilobatin, quercetin, luteolin, naringenin, apigenin, phloretin, kaempferol, baicalein, and wogonin were weighed and dissolved in methanol to obtain the reference solution with the final concentrations of 10.2, 9.8, 10.0, 9.8, 9.9, 10.5, 10.2, 10.8, 9.3, 10.1, 10.6, 9, 4, and 10.5 ug/mL, respectively, and then these solutions were stored in −4°C before analysis.
Chromatography analysis was performed on an Ultimate 3000 focused system (Dionex, Sunnyvale, CA, USA) consisting of an online vacuum degasser, a binary pump, and an autosampler. The sample separation was carried out on the Hypersil GOLD C18 column (100 × 2.1 mm, 1.9
All LC-MSn analyses were performed on the Q-Exactive Focus Orbitrap MS connected to the UHPLC system via a heated electrospray ionization source (Thermo Electron, Bremen, Germany). The optimized tune operating parameters in negative ion mode were listed as follows: sheath gas and auxiliary gas flow rate of 30 and 10 arbitrary, respectively; the capillary and auxiliary gas heater temperatures of 320°C and 350°C, respectively; spray voltage of 3.0 kV; RF lens of 50; High-resolution MS analysis was performed at full scan MS1 with the mass range of m/z 100–1000 at a resolution of 35000 and targeted MS2 at a resolution of 17500 triggered by parallel reaction monitoring mode; nitrogen was set as sheath, auxiliary, and collision gas; the isolation widow was 2 amu, and the normalized collision energy (NCE) was 30%.
All high-resolution MS data were acquired and processed using the Xcalibur version (2.0 software, Thermo Fisher Scientific, San Jose, CA, USA). The compounds were detected by the Compound Discover version 3 using the metabolism workflow templates by the expected compounds predicted method [
TCMSP database is a free and online database for potential target identification of small molecules, especially TCM. The target genes were converted to the official gene symbol by STRING (
STRING was a free tool, which can construct the PPI network by uploading the potential targets. The species was set as “
GeneMANIA (
The GO and KEGG pathway analysis was performed on the DAVID (
In order to identify flavonoids fully, an analytical strategy based on UHPLC Q-Exactive Focus Orbitrap MS was established in this study. First, the sample was prepared and injected into the UHPLC Q-Exactive Focus Orbitrap MS to gain the full scan high-resolution MS data. Then, those data were processed using Compound Discover software with metabolism workflow to predict and detect the molecule of flavonoids. Third, the MS2 of the predicted molecule were acquired using UHPLC Q-Exactive Focus Orbitrap MS by parallel reaction monitoring mode. Finally, the compounds were identified based on the full scan MS, MS2 data, retention time, and bibliography.
The total content of flavonoids was measured by NaNO2-Al(NO3)3-NaOH spectrophotometric colorimetry [
The retention time and mass spectrometric data of flavone in
Peak | Theoretical mass, m/z | Experimental mass, m/z | Error (ppm) | Formula | MS/MS fragment | Identification | |
---|---|---|---|---|---|---|---|
1 | 3.59 | 465.1038 | 465.1039 | 0.21 | C21H22O12 | MS2[465]: 285.0406(100), 125.0233(72), 275.0563(45), 177.0185(29), 303.0512(20), 151.0032(17) | Taxifolin-glucoside |
2 | 3.72 | 465.1038 | 465.1033 | −1.08 | C21H22O12 | MS2[465]: 285.0405(100), 125.0233(38), 273.0043(28), 177.0185(19), 303.0511(15) | Taxifolin-glucoside |
3 | 3.88 | 465.1038 | 465.1042 | 0.86 | C21H22O12 | MS2[465]: 285.0405(100), 125.0233(36), 177.0184(18), 275.0566(12), 303.0512(10) | Taxifolin-glucoside |
4 | 4.13 | 463.0882 | 463.0881 | −0.22 | C21H20O12 | MS2[463]: 287.0562(100), 259.0612(78), 125.0232(32) | Eriodictyol-glucuronide |
5 | 4.19 | 449.1089 | 449.1093 | 0.89 | C21H22O11 | MS2[449]: 259.0611(100), 287.0561(33), 178.9979(15), 125.0231(13) | Eriodictyol-glucoside |
6 | 4.41 | 593.1512 | 593.1535 | 3.88 | C27H30O15 | MS2[593]: 353.0667(100), 383.0774(54), 473.1092(35), 125.0234(25), 413.0874(6) | Vicenin II |
7 | 4.56 | 449.1089 | 449.1092 | 0.67 | C21H22O11 | MS2[449]: 259.0612(100), 269.0455(76), 287.0563(44), 125.0233(38), 178.9977(22) | Eriodictyol-glucoside |
8 | 5.61 | 465.1038 | 465.1042 | 0.86 | C21H22O12 | MS2[465]: 125.0233(100), 285.0405(62), 259.0610(48), 275.0566(32), 303.0512(18) | Taxifolin-glucoside |
9 | 5.75 | 479.0831 | 479.0834 | 0.63 | C21H20O13 | MS2[479]: 285.0405(100), 125.0233(41), 303.0512(18), 177.0186(17), 169.0133(15), 259.0613(14) | Taxifolin-glucuronide |
10 | 5.88 | 449.1089 | 449.1092 | 0.67 | C21H22O11 | MS2[449]: 269.0456(100), 151.0026(69), 178.9979(39), 125.0231(13), 259.0612(12), 287.0562(10) | Eriodictyol-glucoside |
11 | 6.12 | 433.1140 | 433.1142 | 0.46 | C21H22O10 | MS2[433]: 271.0614(100), 151.0027(73). 119.0491(18), 125.0229(8) | Naringenin-glucoside |
12 | 6.22 | 625.1408 | 625.1408 | 0.00 | C27H30O17 | MS2[625]: 287.0558(100), 113.0231(40), 151.0030(38) | Eriodictyol-glucoside-glucuronide |
13 | 6.57 | 303.0510 | 303.0507 | −0.99 | C15H12O7 | MS2[303]: 125.0234(100), 285.0407(38) | |
14 | 6.59 | 433.1140 | 433.1141 | 0.23 | C21H22O10 | MS2[433]: 271.0613(100), 151.0027(72), 119.0491(16) | Naringenin-glucoside |
15 | 6.85 | 463.0882 | 463.0886 | 0.86 | C21H20O12 | MS2[463]: 300.0272(100), 301.0350(70) | Hyperoside |
16 | 6.90 | 449.1089 | 449.1094 | 1.11 | C21H22O11 | MS2[449]:151.0028(100), 287.0562(73), 135.0442(32) | Eriodictyol-glucoside |
17 | 7.18 | 463.0882 | 463.0886 | 0.86 | C21H20O12 | MS2[463]: 300.0279(100), 301.0358(50), 151.0025(8), 178.9976(7) | Isoquercitrin |
18 | 7.75 | 597.1825 | 597.1832 | 1.17 | C27H34O15 | MS2[597]: 357.0983(100), 387.1087(85), 315.0882(22), 417.1165(21), 358.1008(15) | Phloretin-C- diglucoside |
19 | 7.99 | 449.1089 | 449.1085 | −0.89 | C21H22O11 | MS2[449]: 167.0340(69), 137.0233(50), 123.0441(12) | Phloretin-glucuronide |
20 | 8.15 | 431.0984 | 431.0992 | 1.86 | C21H20O10 | MS2[431]: 269.0445(100) | Baicalein-glucoside |
21 | 9.60 | 431.0984 | 431.0988 | 0.93 | C21H20O10 | MS2[431]: 269.0443(100) | Baicalein-glucoside |
22 | 9.61 | 449.1089 | 449.1086 | −0.67 | C21H22O11 | MS2[449]: 151.0026(100), 287.0563(65), 135.0442(22) | Eriodictyol-glucoside |
23 | 9.68 | 565.1563 | 565.1568 | 0.88 | C26H30O14 | MS2[565]: 271.0614(100), 151.0026(48) | Naringenin-pentoside-glucoside |
24 | 9.82 | 433.1140 | 433.1142 | 0.46 | C21H22O10 | MS2[433]: 271.0615(100), 151.0028(41), 119.0491(14) | Naringenin-glucoside |
25 | 10.38 | 431.0984 | 431.0992 | 1.86 | C21H20O10 | MS2[431]: 268.0369(100), 269.0444(47) | Apigenin-7-glucoside |
26 | 11.18 | 463.0882 | 463.0886 | 0.86 | C21H20O12 | MS2[463]: 151.0027(100), 113.0231(66), 287.0562(52), 161.0234(40), 337.0569(39), 135.0439(23) | Eriodictyol-glucuronide |
27 | 12.16 | 435.1297 | 435.1300 | 0.69 | C21H24O10 | MS2[435]: 273.0769(100), 167.0340(69), 125.00232(9), 179.0341(6) | Phlorizin |
28 | 12.36 | 477.0674 | 477.0678 | 0.84 | C21H18O13 | MS2[477]: 301.0353(100), 151.0029(11), 178.9983(9) | Quercetin-glucuronide |
29 | 12.51 | 431.0984 | 431.0986 | 0.46 | C21H20O10 | MS2[431]: 268.0371(100), 269.0441(18), 239.0337(9) | Apigenin-4′-glucoside |
30 | 12.62 | 477.0674 | 477.0678 | 0.84 | C21H18O13 | MS2[477]: 301.0354(100), 178.9983(12), 151.0029(8) | Quercetin-glucuronide |
31 | 13.00 | 433.1140 | 433.1144 | 0.92 | C21H22O10 | MS2[433]: 271.0612(100), 151.0027(46) | Naringenin-glucoside |
32 | 13.39 | 567.1719 | 567.1730 | 1.94 | C26H32O14 | MS2[567]: 273.0771(100), 167.0340(42), 125.0020(22) | Phloretin-pentoside-glucoside |
33 | 13.45 | 287.0561 | 287.0564 | 1.05 | C15H12O6 | MS2[287]: 151.0027(100), 135.0441(69), 107.0125(10) | Eriodictyol |
34 | 13.57 | 449.1089 | 449.1085 | −0.89 | C21H22O11 | MS2[449]: 137.0233(100), 167.0340(89), 123.0440(8) | Phloretin-glucuronide |
35 | 13.67 | 435.1297 | 435.1303 | 1.38 | C21H24O10 | MS2[435]: 273.0767(100), 167.0339(52) | Trilobatin |
36 | 13.86 | 301.0354 | 301.0356 | 0.66 | C15H10O7 | MS2[301]: 151.0027(100), 178.9978(66), 121.0284(21), 107.0125(8) | Quercetin |
37 | 13.95 | 285.0405 | 285.0407 | 0.70 | C15H10O6 | MS2[285]: 285.0404(100), 151.0033(12), 133.0290(11), 175.0395(8) | Luteolin |
38 | 14.79 | 271.0611 | 271.0614 | 1.11 | C15H12O5 | MS2[271]: 151.0027(100), 119.0499(36), 177.0184(14), 93.0333(14), 107.0126(9) | Naringenin |
39 | 14.95 | 269.0455 | 269.0454 | −0.37 | C15H10O5 | MS2[269]: 269.0455(100), 117.0344(8), 149.0239(7), 151.0035(7) | Apigenin |
40 | 14.99 | 273.0768 | 273.0771 | 1.10 | C15H14O5 | MS2[273]: 167.0341(100), 123.0438(18), 119.0491(12),125.0231(8) | Phloretin |
41 | 15.07 | 285.0405 | 285.0405 | 0.00 | C15H10O6 | MS2[285]: 285.0403(100), 151.0030(5) | Kaempferol |
42 | 15.49 | 269.0455 | 269.0453 | −0.74 | C15H10O5 | MS2[269]: 269.0453(100), 241.0504(9), 251.0348(8), 223.0399(7) | Baicalein |
43 | 16.71 | 283.0612 | 283.0613 | 0.35 | C16H12O5 | MS2[283]: 268.0376(100), 163.0035(7) | Wogonin |
The high-resolution extracted ion chromatogram (HREIC) in 5 ppm for the multiple compounds in
Peaks 15, 17, 27, 33, 35–43 were unanimously identified as hyperoside, isoquercitrin, phlorizin, eriodictyol, trilobatin, quercetin, luteolin, naringenin, apigenin, phloretin, kaempferol, baicalein, and wogonin, respectively, by comparing the retention time, high-resolution mass measurement, and MS2 spectrum with those reference standards.
Peak 13 was eluted at 6.57 min and possessed the deprotonated ion [M−H]− at m/z 303.0507 (−0.99 ppm, C15H11O7). The fragment ions at m/z 125.0234 (−8.14 ppm, C6H5O3) and 285.0407 (0.84 ppm, C15H9O6) were detected in the MS2 spectrum, which is consistent with the MS data of taxifolin in bibliography [
Peaks 4 and 26 were eluted at 4.13 and 11.18 min, respectively. All of them showed the same deprotonated ion [M−H]− at m/z 463.088 (C21H19O12), 176.032 Da(C6H8O6, glucuronide moiety) more than that of eriodictyol, suggesting they are eriodictyol-glucuronide, which were further identified by the presence of fragmentation ion at m/z 287.056 (C15H11O6). In a similar way, peaks 9, 28, and 30 were tentatively identified as taxifolin-glucuronide, quercetin-glucuronide, and quercetin-glucuronide, respectively.
Peaks 5, 7, 10, 16, 19, 22, and 34 were eluted at 4.19, 4.56, 5.88, 6.90, 7.99, 9.61, and 13.57 min, with the same deprotonated ion [M−H]− at m/z 449.109 (C21H21O11). Peaks 19 and 34 possessed the fragment ions at m/z 167.034 (C8H7O4) and m/z 123.044 (C7H7O2), which are the diagnosis fragmentation ions of phloretin, suggesting they were phloretin derivatives. Thus, Peaks 19 and 34 were tentatively inferred as phloretin-glucuronide. Peaks 5, 7, 10, 16, and 22 yielded the same fragmentation ion at m/z 287.056 (C15H11O6), suggesting they were eriodictyol derivatives. The ion at m/z 287.056 was yielded by the neutral loss of 162.053 (C6H10O5, glucose moiety), suggesting the presence of glucose moiety. Therefore, they were tentatively characterized as eriodictyol-glucoside.
Peak 6 with the deprotonated ion [M−H]− at m/z 593.1535 (3.88 ppm, C27H29O15) was eluted at 4.41 min. It yielded fragment ions at m/z 353.0667 (0.07 ppm, C19H13O7), 383.0774 (0.42 ppm, C20H15O8), 473.1092(0.56 ppm, C23H21O11), and 413.0874 (−0.98 ppm, C21H17O9), resulting from the loss of C4H8O4 + C4H8O4, C4H8O4 + C3H6O3, C4H8O4, and C3H6O3 + C3H6O3, respectively, suggesting the presence of two carbon-glucoside. According to the published paper [
Peaks 11, 14, 24, and 31 generated the same quasimolecular ion [M−H]− at m/z 433.114 (C21H21O10), 162 Da(C6H10O5, glucose moiety) more than that of naringenin (peak 38), suggesting they were naringenin-glucoside, which were further confirmed by the presence of m/z 271.061 and 151.003 in MS2 spectrum.
Peak 12 eluted at 6.22 min and showed a pseudomolecular ion at m/z 625.1408 (0.00 ppm, C27H29O17), 176.032 Da(C6H8O6, glucuronide moiety) more than that of eriodictyol-glucoside, suggesting it is eriodictyol-glucoside-glucuronide, which was confirmed by the presence of the base peak at m/z 287.0558 (eriodictyol).
Peaks 20, 21, 25, and 29 presented the same deprotonated ion [M−H]− at m/z 431.099 (C21H19O10) and generated the same fragment ions at m/z 269.044 (C15H9O5) by loss of the glucose moiety (C6H10O5), which suggested the presence of glucose moiety. The base peak at m/z 268.037 [Y0–H] ions in the MS2 spectrum of peaks 25 and 29 was a characteristic of apigenin aglycone. According to the published paper [
Peak 32 was detected at 13.39 min. It presented a pseudomolecular ion at m/z 567.1730 (1.94 ppm, C26H32O14) and exhibited the MS2 fragmentation ions at m/z 273.0771 (1.10 ppm, C15H13O5), resulting from the loss of glucose moiety and pentoside moiety (294.096). Thus, peak 32 was tentatively characterized as phloretin-pentoside-glucoside. In a similar way, peak 23 was tentatively identified as naringenin-pentoside-glucoside.
212 putative targets of flavonoids were obtained from the TCMSP database. A visual compounds-targets network with 224 nodes and 440 edges was built by Cytoscape Version 3.7.2 (Figure
In order to find the key targets of flavonoids, a total of 212 putative targets were imported into the STRING to obtain the protein-protein interaction (PPI) data. The PPI network with 206 nodes and 3980 edges was established by Cytoscape (Figure
Among the 23 key target genes and their interacting genes, it was found that 42.75 % had coexpression characteristics, 41.10 % displayed physical interactions characteristic. Other characteristics, including pathway, genetic interactions, colocalization, and shared protein domains, are displayed in Figure
Protein network of core target genes by GeneMANIA. Black nodes represent target proteins, and connecting colors indicate different correlations.
In order to further study the 23 core target genes, GO and KEGG pathway analysis were performed by DAVID. GO term enrichment analysis results were divided into the biological process (BP, 23/23), cell compound (CC, 23/23), and molecular function (MF, 23/23). A total of 158 BP, 16 CC, and 33 MF has a
GO analysis by DAVID: red represents the biological process, yellow represents the cellular component, green represents the molecular function.
KEGG pathway analysis.
Based on the target and KEGG pathway analysis, the entire compounds, targets, and pathway network were established by Cytoscape. The network with 122 nodes and 595 edges is shown in Figure
Chemical-target genes-pathway network: red represents chemical; blue represents target genes; green represents pathway.
In the present investigation, this finding revealed that
The data used to support the finding of this study are available from the corresponding author upon request.
The authors declare no conflicts of interest.
Zaiqi Zhang and Wei Cai designed the study. Kailin Li, Wei Cai, Shihan Qin, Pei Xiong, Jie Peng, and Silin Shi performed the experiments. Kailin Li, Wei Cai and Shihan Qin performed data analyses. Kailin Li and Wei Cai wrote the experiments; Zaiqi Zhang, Shihan Qin, Pei Xiong, Jie Peng, and Silin Shi critically reviewed the manuscript. All authors read and approved the final manuscript. Wei Cai and Kailin Li contributed equally to this work.
This work was supported by the National Natural Science Foundation of China (no. 81603393), Natural Science Foundation of Hunan Province (no. 2018JJ3376), the Scientific Research Fund of Hunan Provincial Education Department (no. 19A353), Hunan Provincial Key Laboratory of Dong Medicine (no. 2015TP1020-03), and the platform construction project of Hunan Provincial key laboratory of Dong medicine (no. 2017CT5025).
Figure S1: compounds-target genes. Figure S2: PPI network. Table S1: compounds-target genes. Table S2: GO analysis. Table S3: KEGG pathway