Identification of prototype compounds and their metabolites in rats' serum from Xuefu Zhuyu Decoction by UPLC-Q-TOF/MS.

In order to remove interfering substances in serum samples and enrich the targeted compounds for more sensitive detection, serum samples were generally pretreated by the methods of protein precipitation, liquid–liquid extraction and solid-phase extraction, respectively ( Feng et al., 2020 , Yan et al., 2019 , Zhang et al., 2019 ). Moreover, addition of a certain amount of acid, the performance of pretreatment can be improved for the acidic compounds. By our study, Cleanert S C18-SPE column (500 mg/3 mL, Bonna-Agela, China) was proved to be suitable as pretreatment method for the serum sample, resulting in the excellent enrichment of the interesting compounds and the satisfactory elimination of interfering signal in MS analysis. The structures of the detected prototype compounds are displayed in Fig. 1 . Fig. 1 Chemical structures of prototype compounds. Chemical structures of prototype compounds. Furthermore, chromatographic separation and MS detection were accomplished to excellently detect the interesting compounds. Acetonitrile – water as the mobile phase system was optimized through adding 0.1% formic acid in water by taking the characteristics of acidic and alkaline analytes into account. Aiming at obtaining comprehensive MS information of analytes as much as possible, UPLC-Q-TOF/MS was performed in positive and negative ion modes. Because of the existence of phenolic hydroxyl and carboxyl groups in the analytes, it is found that the negative ion mode has good MS response. Illumination of prototype compounds absorbed into blood is the prerequisite issue to clarify the exposure of chemical compounds in vivo from Chinese materia medica (CMM), which will pave the way for identification of metabolites derived from these prototype compounds. Through our previous study, flavanones, saponins and phenolic acids were proven to be the representative compounds in XFZYD ( Zhang et al., 2012 , Zhang et al., 2015 ). In our study, characterization of prototype compounds was performed from the dosed serum by analyzing the blank serum, dosed serum, and reference standards. The extracted ion chromatograms of prototype compounds from the serum samples and mixed standards are shown in Fig. 2 . Compared the MS data of the serum samples with those of authentic compounds ( Table 1 and Supplementary Table S1 ), 24 prototype compounds were characterized, including 13 flavanones (HSYA, liquiritin, isoquercitrin, narirutin, naringin, rhoifolin, hesperidin, neohesperidin, liquiritigenin, naringenin, isoliquiritigenin, formononetin and nobiletin), four saponins (platycodin D, ginsenoside-Ro, 18 β -glycyrrhizic acid and saikosaponin A), three phenolic acids (PHBA, PHCA, and ferulic acid), and four others ( β -ecdysterone, amygdalin, albiflorin and paeoniflorin). Fig. 2 Extracted ion chromatograms of prototype compounds from serum sample solution and mixed standards solution (Blank serum sample (A), dosed serum sample (B), and mixed standards sample (C) in negative mode; Blank serum sample (D), dosed serum sample (E), and mixed standards sample (F) in positive mode. Table 1 MS data of prototype compounds from XFZYD in serum by UPLC-Q-TOF/MS. No. Compounds t R (min) Formula Negative ion mode (−) Positive ion mode (+) [M−H] − , ( m / z ) [M+H] + , ( m / z ) P1 PHBA 4.080 C 7 H 6 O 3 137.0239 – P2 HSYA 5.512 C 27 H 32 O 16 611.1612 613.1769 P3 Amygdalin 6.200 C 20 H 27 NO 11 456.1506 458.1662 P4 PHCA 7.532 C 9 H 8 O 3 163.0395 – P5 Albiflorin 7.922 C 23 H 28 O 11 525.1608 * 481.1710 P6 Paeoniflorin 8.739 C 23 H 28 O 11 525.1608 * 498.1985 # P7 Ferulic acid 9.182 C 10 H 10 O 4 193.0501 – P8 Liquiritin 10.355 C 21 H 22 O 9 417.1186 419.1342 P9 Isoquercitrin 10.698 C 21 H 20 O 12 463.0877 465.1033 P10 Narirutin 12.73 C 27 H 32 O 14 579.1714 581.1870 P11 β -Ecdysterone 12.814 C 27 H 44 O 7 479.3009 481.3165 P12 Naringin 13.474 C 27 H 32 O 14 579.1714 581.1870 P13 Rhoifolin 13.711 C 27 H 30 O 14 577.1557 579.1714 P14 Hesperidin 14.372 C 28 H 34 O 15 609.1819 611.1976 P15 Neohesperidin 15.112 C 28 H 34 O 15 609.1819 611.1976 P16 Liquiritigenin 15.521 C 15 H 12 O 4 255.0657 257.0814 P17 Naringenin 18.424 C 15 H 12 O 5 271.0606 273.0763 P18 Platycodin D 19.254 C 57 H 92 O 28 1223.5697 1225.5853 P19 Isoliquiritigenin 20.214 C 15 H 12 O 4 255.0657 257.0814 P20 Formononetin 20.776 C 16 H 12 O 4 267.0657 269.0814 P21 Ginsenoside-Ro 21.398 C 48 H 76 O 19 955.4903 957.5059 P22 18β-Glycyrrhizic acid 22.356 C 42 H 62 O 16 821.3960 823.4116 P23 Nobiletin 23.130 C 21 H 22 O 8 – 403.1393 P24 Saikosaponin A 24.174 C 42 H 68 O 13 779.4582 781.4738 P = prototype compound, “−” undetected, “ * ” [M−H+HCOOH] − , “ # ” [M+NH 4 ] + . Extracted ion chromatograms of prototype compounds from serum sample solution and mixed standards solution (Blank serum sample (A), dosed serum sample (B), and mixed standards sample (C) in negative mode; Blank serum sample (D), dosed serum sample (E), and mixed standards sample (F) in positive mode. MS data of prototype compounds from XFZYD in serum by UPLC-Q-TOF/MS. P = prototype compound, “−” undetected, “ * ” [M−H+HCOOH] − , “ # ” [M+NH 4 ] + . Studies have shown that the prototype compounds identified in XFZYD play an important role in the prevention and treatment of cardiovascular and cerebrovascular diseases. For instance, flavonoids can exert vasodilatory effects in vitro by regulating eNOS or Ca 2+ channels ( Tang et al., 2021 ). Phenolic acids protect the blood brain barrier through MMP-9 inhibition and anti-inflammation ( Zhang & Song et al., 2018 ). Saponins have the pharmacological effects of anti-atherosclerosis, myocardial protection, and anti-thrombosis ( Li et al., 2015 ). For example, HSYA, as the marker compound of safflower, has been widely used for the treatment of cerebrovascular and cardiovascular diseases due to its property of promoting blood circulation and removing blood stasis ( Yang et al., 2020 ). Formononetin could alleviate the development of atherosclerosis by regulating the interaction between KLF4 and SRA ( Ma et al., 2020 ). Amygdalin was reported to have the effect of reducing the development of atherosclerosis by inhibiting inflammatory reaction and promoting the immune regulation function of T cells ( Deng et al., 2011 ). Naringin, a major compound of flavanone, which can be extracted from many CMM herbs, has the effects of anti-atherosclerosis, anti-hypertension, and myocardial protection ( Hsueh et al., 2016 , Sun et al., 2019 ). Isoliquiritigenin is a flavonoid compound from Glycyrrhiza glabra that has been proven to attenuate atherosclerosis lesion and decrease blood lipid level by inhibiting TRPC5 channel ( Qi et al., 2020 ). Also, it has been reported that hesperidin prevents the redox imbalance induced by hyperlipidemia ( Kumar, Akhtar, & Rizvi, 2020 ). In vitro and in vivo experiments showed that ferulic acid has antithrombotic, hypolipidemic activities, and so on ( Choi et al., 2018 , Kamal-Eldin et al., 2000 , Zhu and Zhang, 2014 ). Enzymes play an important role in the metabolism of CMM compounds. It can metabolize prototype compounds derived from the studied formula into active metabolites, contributing to the treatment of diseases. β -Glucosidase is abundant enzyme in intestinal, which can easily transform glycoside into aglycone by deglycosylation reaction ( Yan et al., 2018 ). In liver, liver microsomal enzymes can metabolize prodrug into metabolites through Ⅰ and Ⅱ phase metabolism ( Zeng et al., 2018 ). By analyzing the chromatograms of the blank serum and dosed serum ( Fig. 3 ), a total of 151 metabolites were tentatively identified by the Metabolyxn (Masslynx V4.1 software) and detailed MS data ( Table 2 and Supplementary Table S2 ). There were 71, 53, 14, and 13 metabolites transformed from flavanones, saponins, phenolic acids and other compounds, respectively. The network was adopted to display the relationship between prototype compounds and metabolites, and 24 prototype compounds (P1–P24) identified from XFZYD have been modified into 151 metabolites, which was shown in Fig. 4 . Interestingly, it can be drawn that one prototype compound can be metabolized into the different metabolites, the different prototype compounds can be metabolized into the same metabolite. M 10-8 taken as a typical example was simultaneously produced by narirutin, naringin, and neohesperidin in vivo . Meanwhile, narirutin, naringin, and neohesperidin can be metabolized into the different metabolites besides M 10-8 , respectively. Fig. 3 Extracted ion chromatograms of metabolites of serum samples. Blank serum sample (A) and dosed serum sample (B) in negative mode; Blank serum sample (C) and dosed serum sample (D) in positive mode. Table 2 MS data and metabolic pathways of metabolites after oral administration of XFZYD by UPLC-Q-TOF/MS. Prototypes Metabolic pathways No. t R (min) Peak No. Negative ion mode (−) Positive ion mode (+) Formula [M−H] − , ( m / z ) [M+H] + , ( m / z ) PHBA Parent P 1 3.99 – 137.0242 – C 7 H 6 O 3 Hydroxylation + Methylation M 1-1 5.74 11 167.0350 – C 8 H 8 O 4 Hydroxylation M 1-2 6.04 14 153.0195 – C 7 H 6 O 4 

 HSYA Parent P 2 5.51 – 611.1616 – C 27 H 32 O 16 Deoxidation M 2-1 11.43 42 595.1682 – C 27 H 32 O 15 

 Amygdalin Parent P 3 6.13 – 456.1506 480.1478 # C 20 H 27 NO 11 Deglycosylation (glucose) M 3-1 7.42 20 294.0985 318.0961 # C 14 H 17 NO 6 M 3-2 7.58 21 294.0984 318.0962 # C 14 H 17 NO 6 Deglycosylation (2 × glucose) + Sulfate conjugation M 3-3 8.90 23 212.0025 – C 8 H 7 NO 4 S 

 PHCA Parent P 4 7.51 – 163.0402 – C 9 H 8 O 3 Hydroxylation + Sulfation conjugation M 4-1 4.90 9 258.9922 – C 9 H 8 O 7 S Sulfate conjugation M 4-2 2.65 2 242.9968 – C 9 H 8 O 6 S M 4-3 4.77 8 242.9970 – C 9 H 8 O 6 S M 4-4 5.84 13 242.9969 – C 9 H 8 O 6 S M 4-5 6.07 15 242.9969 – C 9 H 8 O 6 S 

 Albiflorin Parent P 5 7.91 – 525.1614 * 503.1537 # C 23 H 28 O 11 De-benzoic acid M 5-1 5.74 12 359.1338 383.1314 # C 16 H 24 O 9 De-benzoic acid + Dehydration M 5-2 10.65 38 341.1243 365.1212 # C 16 H 22 O 8 De-benzoic acid + Acetylation M 5-3 10.80 40 401.1451 425.1432 # C 18 H 26 O 10 

 Paeoniflorin Parent P 6 8.72 – 525.1609 * 503.1525 # C 23 H 28 O 11 De-debenzoylpaeoniflorin + Glucuronide conjugation M 6-1 2.38 1 343.0662 * – C 13 H 14 O 8 De-debenzoylpaeoniflorin + Hydroxylation + sulfate conjugation M 6-2 2.73 3 262.9869 * – C 7 H 6 O 6 S De-debenzoylpaeoniflorin + Sulfate conjugation M 6-3 3.06 4 246.9924 * – C 7 H 6 O 5 S 

 Ferulic acid Parent P 7 9.18 – 193.0507 – C 10 H 10 O 4 Glucuronide conjugation M 7-1 4.34 6 369.0823 – C 16 H 18 O 10 M 7-2 6.57 18 369.0834 – C 16 H 18 O 10 Sulfate conjugation M 7-3 3.13 5 273.0078 – C 10 H 10 O 7 S M 7-4 5.38 10 273.0081 – C 10 H 10 O 7 S M 7-5 6.41 16 273.0080 – C 10 H 10 O 7 S M 7-6 6.56 17 273.0080 – C 10 H 10 O 7 S Demethylation + Sulfate conjugation M 7-7 4.90 9 258.9918 – C 9 H 8 O 7 S Demethylation + Deoxidation + Sulfate conjugation M 7-8 2.65 2 242.9968 – C 9 H 8 O 6 S M 7-9 4.66 7 242.9973 – C 9 H 8 O 6 S M 7-10 4.77 8 242.9970 – C 9 H 8 O 6 S M 7-11 6.07 15 242.9968 – C 9 H 8 O 6 S 

 Liquiritin Parent P 8 10.32 – 417.1195 419.1353 C 21 H 22 O 9 Glucuronide conjugation M 8-1 6.61 19 593.1511 – C 27 H 30 O 15 M 8-2 10.19 31 593.1508 595.1645 C 27 H 30 O 15 Hydroxylation + Dehydrogenation M 8-3 10.04 25 431.0979 433.1131 C 21 H 20 O 10 M 8-4 10.36 35 431.0979 433.1134 C 21 H 20 O 10 Deglycosylation (glucose) + Hydroxylation + Glucuronide conjugation M 8-5 13.11 54 447.0931 449.1082 C 21 H 20 O 11 Hydroxylation + Methylation M 8-6 17.54 80 447.1289 449.1444 C 22 H 24 O 10 Hydroxylation + Sulfation conjugation M 8-7 14.31 63 513.0705 – C 21 H 22 O 13 S Deglycosylation (glucose) + Methylation M 8-8 18.62 88 269.0817 271.0978 C 16 H 14 O 4 Deglycosylation (glucose) + Sulfate conjugation M 8-9 12.41 49 335.0229 337.0376 C 15 H 12 O 7 S M 8-10 18.14 85 335.0230 337.0384 C 15 H 12 O 7 S 

 Isoquercitrin Parent P 9 10.69 – 463.0880 465.102 C 21 H 20 O 12 Methylation + Deoxidation M 9-1 13.98 56 461.1090 – C 22 H 22 O 11 M 9-2 14.15 60 461.1086 – C 22 H 22 O 11 M 9-3 18.98 93 461.1088 – C 22 H 22 O 11 Deoxidation M 9-4 15.33 66 447.0936 – C 21 H 20 O 11 Deoxidation + Sulfate conjugation M 9-5 10.11 30 527.0497 – C 21 H 20 O 14 S M 9-6 10.26 33 527.0498 – C 21 H 20 O 14 S 

 Narirutin Parent P 10 12.72 – 579.1718 581.1868 C 27 H 32 O 14 Deglycosylation (rhamnose) + Dehydrogenation M 10-1 10.04 26 431.0979 433.1131 C 21 H 20 O 10 M 10-2 10.36 36 431.0979 433.1134 C 21 H 20 O 10 Deglycosylation (rutinose) + Glucuronide conjugation M 10-3 13.11 54 447.0931 449.1082 C 21 H 20 O 11 Hydroxylation M 10-4 10.72 39 595.1658 597.181 C 27 H 32 O 15 M 10-5 11.43 43 595.1677 – C 27 H 32 O 15 Hydroxylation + Dehydrogenation M 10-6 18.11 82 593.1506 595.1644 C 27 H 30 O 15 Hydroxylation + Methylation M 10-7 17.23 79 609.1833 611.1979 C 28 H 34 O 15 Deglycosylation (rutinose) + Hydroxylation + Methylation M 10-8 13.04 52 301.0722 303.0875 C 16 H 14 O 6 M 10-9 13.23 55 301.0717 303.0872 C 16 H 14 O 6 M 10-10 14.23 62 301.0722 303.0883 C 16 H 14 O 6 Methylation M 10-11 14.01 58 593.1874 – C 28 H 34 O 14 

 β -Ecdysterone Parent P 11 12.82 – 525.3080 * – C 27 H 44 O 7 Methylation M 11-1 22.44 100 539.3206 * – C 28 H 46 O 7 Reduction + Deoxidation M 11-2 22.48 101 511.3267 * – C 27 H 46 O 6 Demethylation + Deoxidation M 11-3 22.63 102 495.2962 * – C 26 H 42 O 6 Acetylation M 11-4 23.97 111 567.3167 * – C 29 H 46 O 8 

 Naringin Parent P 12 13.47 – 579.1719 581.1863 C 27 H 32 O 14 Deglycosylation (rhamnose) + Dehydrogenation M 12-1 10.04 26 431.0979 433.1131 C 21 H 20 O 10 M 12-2 10.36 36 431.0979 433.1134 C 21 H 20 O 10 Deglycosylation (neohesperidose) + Glucuronide conjugation M 12-3 13.11 54 447.0931 449.1082 C 21 H 20 O 11 Hydroxylation + Dehydrogenation M 12-4 18.11 83 593.1506 595.1644 C 27 H 30 O 15 Deglycosylation (neohesperidose) + Hydroxylation + Methylation M 12-5 13.04 52 301.0722 303.0875 C 16 H 14 O 6 Methylation M 12-6 14.01 59 593.1874 – C 28 H 34 O 14 

 Rhoifolin Parent P 13 13.71 – 577.1590 579.1697 C 27 H 30 O 14 Deglycosylation (neohesperidose) + Hydroxylation + Glucuronide conjugation M 13-1 12.33 48 461.0726 463.0879 C 21 H 18 O 12 Deglycosylation (rhamnose) + Hydroxylation + Methylation M 13-2 14.15 61 461.1086 – C 22 H 22 O 11 M 13-3 13.98 57 461.1090 – C 22 H 22 O 11 M 13-4 18.98 94 461.1088 – C 22 H 22 O 11 M 13-5 19.10 95 461.1084 – C 22 H 22 O 11 Deglycosylation (neohesperidose) + Hydroxylation + Sulfation conjugation M 13-6 9.86 24 364.9967 – C 15 H 10 O 9 S Deglycosylation (rhamnose) + Hydroxylation + Sulfation conjugation M 13-7 10.09 28 527.0500 – C 21 H 20 O 14 S M 13-8 10.25 32 527.0500 – C 21 H 20 O 14 S Deglycosylation (rhamnose) + Hydroxylation M 13-9 15.33 67 447.0936 – C 21 H 20 O 11 Deglycosylation (rhamnose) + Methylation M 13-10 18.64 91 445.1133 447.1294 C 22 H 22 O 10 Deglycosylation (rhamnose) M 13-11 10.04 26 431.0979 433.113 C 21 H 20 O 10 M 13-12 10.36 36 431.0979 – C 21 H 20 O 10 M 13-13 15.62 72 – 433.1136 C 21 H 20 O 10 M 13-14 16.30 77 – 433.113 C 21 H 20 O 10 Deglycosylation (rhamnose) + Reduction M 13-15 11.57 45 433.1135 – C 21 H 22 O 10 M 13-16 17.09 78 – 435.129 C 21 H 22 O 10 Deglycosylation (rhamnose) + Sulfate conjugation M 13-17 12.54 51 511.0547 – C 21 H 20 O 13 S M 13-18 8.22 22 511.0553 – C 21 H 20 O 13 S 

 Hesperidin Parent P 14 14.37 – 609.1814 611.1979 C 28 H 34 O 15 Deglycosylation (rutinose) + 2 × Glucuronide conjugation M 14-1 11.23 41 653.1345 – C 28 H 30 O 18 Glucuronide conjugation M 14-2 11.60 46 785.2145 787.2285 C 34 H 42 O 21 Deglycosylation (rhamnose) + Dehydrogenation M 14-3 14.15 61 461.1087 463.1242 C 22 H 22 O 11 Deglycosylation (rutinose) + Glucuronide conjugation M 14-4 15.03 65 477.1036 479.1185 C 22 H 22 O 12 Dehydrogenation + Demethylation M 14-5 18.11 82 593.1506 595.1644 C 27 H 30 O 15 

 Neohesperidin Parent P 15 15.09 – 609.1842 611.1967 C 28 H 34 O 15 Deglycosylation (neohesperidose) + 2 × Glucuronide conjugation M 15-1 11.23 41 653.1345 – C 28 H 30 O 18 Demethylation M 15-2 11.43 43 595.1677 597.1825 C 27 H 32 O 15 Deglycosylation (rhamnose) + Dehydrogenation M 15-3 14.15 61 461.1087 463.1241 C 22 H 22 O 11 Glucuronide conjugation M 15-4 11.60 47 785.2145 787.2269 C 34 H 42 O 21 Deglycosylation (neohesperidose) + Glucuronide conjugation M 15-5 15.03 65 477.1036 479.1185 C 22 H 22 O 12 M 15-6 15.47 70 477.1034 – C 22 H 22 O 12 Deglycosylation (neohesperidose) M 15-7 13.04 52 301.0722 303.0876 C 16 H 14 O 6 M 15-8 13.23 55 301.0717 303.087 C 16 H 14 O 6 M 15-9 14.23 62 301.0722 303.0882 C 16 H 14 O 6 Deoxidation M 15-10 14.01 59 593.1874 – C 28 H 34 O 14 Deglycosylation (rhamnose) + Sulfate conjugation M 15-11 14.58 64 543.0828 – C 22 H 24 O 14 S 

 Liquiritigenin Parent P 16 15.51 – 255.0659 257.0822 C 15 H 12 O 4 Glucuronide conjugation M 16-1 10.04 27 431.0979 433.1131 C 21 H 20 O 10 M 16-2 10.36 37 431.0979 433.1134 C 21 H 20 O 10 Hydroxylation M 16-3 15.38 68 271.0621 – C 15 H 12 O 5 Hydroxylation + Glucuronide conjugation M 16-4 10.09 29 447.0932 – C 21 H 20 O 11 M 16-5 13.11 54 447.0931 449.1082 C 21 H 20 O 11 Methylation M 16-6 18.62 88 269.0817 271.0978 C 16 H 14 O 4 Sulfate conjugation M 16-7 11.55 44 335.0225 337.0385 C 15 H 12 O 7 S M 16-8 12.47 50 335.0229 337.0376 C 15 H 12 O 7 S 

 Naringenin Parent P 17 18.42 – 271.0602 273.0771 C 15 H 12 O 5 Glucuronide conjugation M 17-1 13.13 54 447.0930 449.1082 C 21 H 20 O 11 Sulfate conjugation M 17-2 15.40 69 351.0183 353.0337 C 15 H 12 O 8 S Deoxidation + Sulfate conjugation M 17-3 18.15 85 335.0230 337.0384 C 15 H 12 O 7 S Methylation + Deoxidation M 17-4 18.62 88 269.0817 271.0978 C 16 H 14 O 4 

 Platycodin D Parent P 18 19.25 – 1223.5681 1225.5792 C 57 H 92 O 28 Deglycosylation (apiose + xylose + rhamnose + arabinose) M 18-1 21.40 98 681.3850 – C 36 H 58 O 12 Deglycosylation (apiose) + Methylation + Deoxidation M 18-2 30.90 142 1089.5479 – C 53 H 86 O 23 

 Isoliquiritigenin Parent P 19 20.21 – 255.0669 257.0823 C 15 H 12 O 4 Hydroxylation + Sulfation conjugation M 19-1 10.27 34 351.0177 353.0334 C 15 H 12 O 8 S Glucuronide conjugation M 19-2 10.36 37 431.0979 433.1134 C 21 H 20 O 10 Sulfate conjugation M 19-3 12.47 50 335.0229 337.0376 C 15 H 12 O 7 S Hydroxylation + Glucuronide conjugation M 19-4 13.11 54 447.0931 449.1082 C 21 H 20 O 11 M 19-5 15.51 71 447.0931 449.1086 C 21 H 20 O 11 Glucuronide conjugation M 19-6 15.62 73 431.0982 433.1137 C 21 H 20 O 10 M 19-7 16.23 76 431.0978 433.113 C 21 H 20 O 10 Sulfate conjugation M 19-8 18.16 86 335.0230 337.0384 C 15 H 12 O 7 S Methylation M 19-9 18.62 89 269.0817 271.0978 C 16 H 14 O 4 

 Formononetin Parent P 20 20.77 – 267.0666 269.0825 C 16 H 12 O 4 Glucuronide conjugation M 20-1 16.11 74 443.0978 445.1135 C 22 H 20 O 10 Hydroxylation + Glucuronide conjugation M 20-2 13.10 53 459.0933 461.108 C 22 H 20 O 11 M 20-3 17.88 81 459.0924 461.108 C 22 H 20 O 11 Reduction M 20-4 18.62 90 269.0817 271.0978 C 16 H 14 O 4 Sulfate conjugation M 20-5 18.68 92 347.0228 349.0386 C 16 H 12 O 7 S 

 Ginsenoside-Ro Parent P 21 21.39 – 955.4894 – C 48 H 76 O 19 Deglycosylation (glucose) + Deglucuronidation + 2 × Dehydrogenation M 21-1 20.69 97 613.3737 615.3871 C 36 H 54 O 8 Deglycosylation (glucose) + Hydroxylation + Dehydrogenation M 21-2 25.43 115 807.4169 – C 42 H 64 O 15 Deglycosylation (glucose) + Deglucuronidation + Hydroxylation + Dehydrogenation M 21-3 30.22 139 631.3857 – C 36 H 56 O 9 

 18 β -Glycyrrhizic acid Parent P 22 22.37 – 821.3950 823.4107 C 42 H 62 O 16 2 × Deglucuronidation + 3 × Hydroxylation M 22-1 23.34 107 517.3167 519.3307 C 30 H 46 O 7 2 × Deglucuronidation + Demethylation + Hydroxylation M 22-2 24.13 112 471.3106 473.3261 C 29 H 44 O 5 M 22-3 25.58 116 471.3114 473.3272 C 29 H 44 O 5 M 22-4 25.73 119 471.3110 473.3272 C 29 H 44 O 5 M 22-5 26.10 122 471.3113 473.327 C 30 H 44 O 5 M 22-6 26.65 125 471.3113 473.3267 C 29 H 44 O 5 M 22-7 30.59 141 471.3115 473.3269 C 29 H 44 O 5 Deglucuronidation + Demethylation + Hydroxylation M 22-8 20.62 96 647.3424 649.3581 C 35 H 52 O 11 2 × Deglucuronidation + Demethylation M 22-9 23.38 108 – 457.332 C 29 H 44 O 4 M 22-10 30.23 140 455.3157 457.3316 C 29 H 44 O 4 M 22-11 31.01 143 455.3166 457.3317 C 29 H 44 O 4 M 22-12 31.31 144 455.3159 457.332 C 29 H 44 O 4 2 × Deglucuronidation + Dehydrogenation M 22-13 28.34 134 – 469.3318 C 30 H 44 O 4 Deglucuronidation + Deoxidation + Dehydrogenation M 22-14 23.88 110 659.3431 661.3585 C 36 H 52 O 11 Hydroxylation M 22-15 18.31 87 837.3928 839.4043 C 42 H 62 O 17 2 × Deglucuronidation + Hydroxylation + Dehydrogenation M 22-16 25.83 121 483.3111 485.3264 C 30 H 44 O 5 M 22-17 27.13 128 483.3103 485.3266 C 30 H 44 O 5 M 22-18 27.99 133 483.3132 485.3272 C 30 H 44 O 5 M 22-19 29.86 136 483.3112 485.3268 C 30 H 44 O 5 Deglucuronidation + Hydroxylation + Dehydrogenation M 22-20 22.96 104 – 661.3583 C 36 H 52 O 11 2 × Deglucuronidation + Hydroxylation + Sulfation conjugation M 22-21 22.93 103 565.2838 567.2986 C 30 H 46 O 8 S M 22-22 23.16 106 565.2840 567.2985 C 30 H 46 O 8 S 2 × Deglucuronidation + Hydroxylation M 22-23 23.85 109 485.3260 487.3422 C 30 H 46 O 5 M 22-24 24.98 114 485.3271 487.3424 C 30 H 46 O 5 M 22-25 25.64 117 485.3273 487.3425 C 30 H 46 O 5 M 22-26 26.12 123 485.3271 487.3426 C 30 H 46 O 5 M 22-27 27.17 129 485.3259 487.3423 C 30 H 46 O 5 M 22-28 27.85 131 485.3275 487.3426 C 30 H 46 O 5 M 22-29 29.87 138 485.3259 487.3424 C 30 H 46 O 5 Deglucuronidation + Hydroxylation M 22-30 21.51 99 661.3586 663.3733 C 36 H 54 O 11 2 × Deglucuronidation M 22-31 32.40 146 469.3322 471.3491 C 30 H 46 O 4 M 22-32 32.69 149 469.3313 471.3472 C 30 H 46 O 4 Deglucuronidation M 22-33 26.81 127 645.3626 647.3785 C 36 H 54 O 10 Deglucuronidation + Sulfate conjugation M 22-34 22.99 105 725.3210 727.3356 C 36 H 54 O 13 S 

 Nobiletin Parent P 23 23.12 – – 403.1392 C 21 H 22 O 8 Hydroxylation M 23-1 16.22 75 – 419.1338 C 21 H 22 O 9 Hydroxylation + Glucuronide conjugation M 23-2 18.11 84 – 595.1641 C 27 H 30 O 15 

 Saikosaponin A Parent P 24 24.17 – 779.4595 – C 42 H 68 O 13 Deglycosylation (glucose + rhamnose) + Dehydrogenation M 24-1 32.40 147 469.3332 471.3491 C 30 H 46 O 4 M 24-2 32.47 148 469.3318 471.3492 C 30 H 46 O 4 M 24-3 32.69 150 469.3313 471.3472 C 30 H 46 O 4 M 24-4 33.15 151 469.3311 471.3474 C 30 H 46 O 4 Deglycosylation (glucose + rhamnose) + Hydroxylation + Dehydrogenation M 24-5 24.89 113 – 487.3425 C 30 H 46 O 5 M 24-6 25.64 118 485.3273 487.3425 C 30 H 46 O 5 M 24-7 26.12 124 485.3271 487.3426 C 30 H 46 O 5 M 24-8 26.70 126 – 487.3424 C 30 H 46 O 5 M 24-9 27.17 130 485.3262 487.3423 C 30 H 46 O 5 M 24-10 27.85 132 485.3257 487.3426 C 30 H 46 O 5 M 24-11 29.18 135 485.3266 487.341 C 30 H 46 O 5 M 24-12 29.86 137 485.3264 487.3424 C 30 H 46 O 5 Deglycosylation (glucose + rhamnose) M 24-13 25.75 120 471.3467 473.3642 C 30 H 48 O 4 Deglycosylation (glucose + rhamnose) + 2 × Dehydrogenation M 24-14 32.31 145 467.3164 469.3309 C 30 H 44 O 4 P = prototype compound, M = metabolite, “−” undetected, “ * ” [M−H+HCOOH] - , “ # ” [M+Na] + . Fig. 4 Metabolic network of prototype compounds in XFZYD. Extracted ion chromatograms of metabolites of serum samples. Blank serum sample (A) and dosed serum sample (B) in negative mode; Blank serum sample (C) and dosed serum sample (D) in positive mode. MS data and metabolic pathways of metabolites after oral administration of XFZYD by UPLC-Q-TOF/MS. P = prototype compound, M = metabolite, “−” undetected, “ * ” [M−H+HCOOH] - , “ # ” [M+Na] + . Metabolic network of prototype compounds in XFZYD. As follows, we introduced the identified metabolites and summarized metabolic pathways of prototype compounds derived from the different characteristics of structures. In this study, 13 prototype flavonoids were identified in rat' serum after oral administration of XFZYD, which were mainly originated from Honghua, Gancao, Zhiqiao, and Danggui. With the aid of Metabolyxn (Masslynx V4.1 software), 71 metabolites from flavonoids were identified ( Table 2 and Supplementary Table S2 ), which were derived from P2 (one metabolite), P8 (10 metabolites), P9 (six metabolites), P10 (11 metabolites), P12 (six metabolites), P13 (18 metabolites), P14 (five metabolites), P15 (11 metabolites), P16 (eight metabolites), P17 (four metabolites), P19 (nine metabolites), P20 (five metabolites) and P23 (two metabolites), respectively. By studying the metabolic pathways in detail, we found that flavonoids mainly undergo glucuronidation, sulfation deglycosylation, demethylation, and deoxidation reactions. For example, neohesperidin (P15) was modified into eleven metabolites (M 15-1 –M 15-11 ) through above-mentioned metabolic pathways ( Fig. 5 ). Fig. 5 Metabolic pathways of neohesperidin. Metabolic pathways of neohesperidin. Glucuronidation: M 15-4 showed a quasi-molecular ion at m / z 785.2145 in the negative ion mode, 176 Da heavier than neohesperidin at m / z 609.1842, suggesting that glucuronidation reaction has happened to neohesperidin. Fragment ions of M 15-4 at m / z 301.0714, 151.0059, 149.0621, and 175.0255 were detected, whose fragmentation characteristics were similar to neohesperidin. The characteristic fragment ion of glucuronic acid was monitored at m / z 175.0255. This result indicated that the occurrence of glucuronidation conjugation to neohesperidin. Deglycosylation + Glucuronidation: In the negative ion mode, M 15-7 , M 15-8 , and M 15-9 showed quasi-molecular ion at m / z 301.0722 and were identified as aglycone of neohesperidin, suggesting that deglycosylation reaction has occurred to neohesperidin by intestine microbiota ( Lin et al., 2020 ). M 15-5 and M 15-6 showed quasi-molecular ion both at m / z 477.1034, indicating that glucuronidation has occurred at the different hydroxyl group of hesperetin. The fragment ions of M 15-5 and M 15-6 were detected at m / z 301.0721, 151.0040 and 149.0617, which were similar to that of neohesperidin. M 15-1 was identified as a product derived from the successive glucuronidation reaction to hesperetin, whose quasi-molecular ion was found at m / z 653.1345. Sulfation: As a sulfated product, M 15-11 showed a quasi-molecular ion at m / z 543.0828 and fragment ions at m / z 463.1240, 301.0385, 151.0044, and 149.0606 in the negative ion mode, indicating that sulfation reaction has happened to the metabolite generated from neohesperidin by losing a rhamnose. Demethylation: M 15-2 was tentatively identified as a demethylated product by comparing the quasi-molecular ion between M 15-2 and neohesperidin, whose gap was 14 Da in the negative ion mode. In addition, M 15-10 was metabolized by neohesperidin through deoxidation reaction by loss of oxygen (16 Da) and showed a quasi-molecular ion at m / z 593.1874 in the negative ion mode. In general, illumination of the proposed metabolic pathway about neohesperidin paved the way for studying the metabolites of other flavonoids. The identified metabolites of flavonoids are displayed in Table 2 and Supplementary Table S2 . Based on the information of chromatographic behavior and MS data, four prototype compounds from saponins primarily derived from Jiegeng, Chuanxiong, Gancao, and Chaihu, were identified by comparing with standards, which were transformed into 53 metabolites, including two metabolites from P18, three metabolites from P21, 34 metabolites from P22, and 14 metabolites from P24 ( Table 2 and Supplementary Table S2 ). Hydrolyzation, dehydrogenation, and hydroxylation were found to happen in the process of saponins metabolism. Give an example about how to identify metabolites, saikosaponin A underwent the mentioned metabolic pathways to produce 14 metabolites (M 24-1 –M 24-14 ) in negative ion mode. Compared to the quasi-molecular ion of saikosaponin A at m / z 779.4595, that of M 24-13 was found to lose 308 Da and detected at m / z 471.3467, indicating that glucose and rhamnose were sequentially hydrolyzed from saikosaponin A to produce aglycone. The fragment ion at m / z 453.3000 was generated from M 24-13 by losing H 2 O. The quasi-molecular ion of M 24-1 –M 24-4 at m / z 469.3318 and the fragment ion at m / z 451.3203 were observed, which was inferred that M 24-1 –M 24-4 were produced by dehydrogenation of M 24-13 . The successive dehydrogenation to M 24-13 produced M 24-14 , whose quasi-molecular ion was monitored at m / z 467.3164 and a fragment ion at m / z 449.3234. The hydroxylation reaction occurred to M 24-1 –M 24-4 , which generated M 24-5 -M 24-12 with the quasi-molecular ion at m / z 485.3264. As phenolic acids derived from Dihuang, Danggui, Chuanxiong, and Jiegeng, P1, P4, and P7 were metabolized into two, five, and 11 metabolites detected in the dosed serum, respectively ( Table 2 and Supplementary Table S2 ). The prominent metabolic reactions were hydroxylation and methylation for the tested compounds. For example, M 1-2 showed a quasi-molecular ion at m / z 153.0195, which was 16 Da more than that of PHBA at m / z 137.0242 and produced the main fragment ions at m / z 109.0293 in the MS spectra, indicating that hydroxylation reaction has happened to PHBA. M 1-1 was metabolized from PHBA by the sequential reaction of hydroxylation (16 Da) and methylation (14 Da), whose quasi-molecular ion was detected at m / z 167.0350. Originated from Niuxi, Taoren, and Chishao, β -ecdysterone, amygdalin, albiflorin, and paeoniflorin were identified and modified into four, three, three, and three metabolites detected in the serum after oral administration of XFZYD ( Table 2 and Supplementary Table S2 ). Methylation, demethylation, acetylation, reduction, glucuronide conjugation, and sulfate conjugation were summarized from the identified metabolites. Taking M 11-1 –M 11-4 metabolized from P11 as an example, M 11-1 provided the quasi-molecular ion at m / z 539.3206, 14 Da more than that of β -ecdysterone at m / z 525.3080, suggesting that methylation reaction has occured. M 11-2 displayed a quasi-molecular ion at m / z 511.3267 and the fragment ions at m / z 351.2148, 333.2022, and 205.0889, showing that reduction and deoxidation reactions have happened to β -ecdysterone. M 11-3 was presumed to be the product of β -ecdysterone through demethylation and deoxidation reactions, whose quasi-molecular ion was monitored at m / z 495.2962. The quasi-molecular ion of M 11-4 was detected at m / z 567.3167, indicating that acetylation reaction has generated to β -ecdysterone.

Acetonitrile and methanol (HPLC grade) were purchased from Fisher Scientific (Fisher, USA). Formic acid and dimethyl sulfoxide were purchased from Meridian Medical Technologies (MREDA, USA). Water for UPLC analysis was purified by a Milli-Q water purification system (Millipore, USA). Reference compounds, including p -hydroxybenzoic acid (PHBA), hydroxysafflor yellow A (HSYA), amygdalin, p -hydroxycinnamic acid (PHCA), ferulic acid, albiflorin, paeoniflorin, liquiritin, isoquercitrin, narirutin, β -ecdysterone, naringin, rhoifolin, hesperidin, neohesperidin, liquiritigenin, naringenin, platycodin D, isoliquiritigenin, formononetin, ginsenoside-Ro, 18 β -glycyrrhizic acid, nobiletin, and saikosaponin A were obtained from the National Institutes for Food and Drug Control (Beijing, China), Tianjin ZhongXin Pharmaceutical Group Co., Ltd. (Tianjin, China), and Top High Bio Technology Co., Ltd. (Nanjing, China). The purity of standards was above 98%. All samples were stored at 4 °C before analysis. As individual standard stock solution, reference compounds were accurately weighed and directly prepared in methanol by using dimethyl sulfoxide as a cosolvent. Then, a mixed standard solution was prepared at about 5 μg/mL. All solution of reference standards was stored at 4 °C. Chromatographic analysis was performed on an ACQUITY™ UPLC system (Waters, Milford, USA) equipped with a binary solvent manager, sample manager and column oven, which was controlled by Masslynx V4.1 software. Chromatographic separation was carried out on an ACQUITY™ UPLC BEH C 18 column (2.1 × 100 mm, 1.7 μm) held at 50 °C. The flow rate was set at 0.3 mL/min. The mobile phase consisted of 0.1% formic acid aqueous solution (A) and acetonitrile (B) in a gradient elution. The program applied was as follows: 0–6.5 min, 3%−11% B; 6.5–15 min, 11%−20% B; 15–20 min, 20%−36% B; 20–27 min, 36%−48% B; 27–30 min, 48%−55% B; 30–33 min, 55%−74% B, and 33–35 min, 74%−90%. The injection volume of the sample solution was 5 μL. The MS analysis was performed by employing a Waters ACQUITY SYNAPT TM G2-S high definition mass spectrometer system (Waters, Milford, USA) equipped with an electrospray ion (ESI) source. The optimal conditions were as follows: capillary voltage at 3.0 and −2.5 kV in positive and negative ion mode, respectively; sampling cone voltage at 40 V; source temperature at 120 °C; desolvation temperature at 400 °C; the flow rate of cone gas and desolvation gas (N 2 ) at 50 L/h and 700 L/h, respectively; the flow rate of collision gas (Ar) at 0.20 mL/min. Data were acquired in centroid mode from m / z 100 to 1500 Da. For accurate mass to charge ratio acquisition, the MS was corrected during data acquisition using a lock mass of leucine-enkephalin (LE) at a concentration of 200 pg/mL via a LockSpray TM interface at a flow rate of 10 μL/min, monitoring the reference ions in the positive ion mode ([M+H] + = 556.2771) and the negative ion mode ([M−H] – = 554.2615) during MS analysis. The decoction pieces (Taoren, Danggui, Chuanxiong, Honghua, Chishao, Dihuang, Zhiqiao, Chaihu, Jiegeng, Niuxi, and Gancao) were purchased from Anguo Oriental Medical Town (Hebei, China) and identified by Professor Tianxiang Li, which were deposited in Tianjin Key Laboratory of TCM Chemistry and Analysis (Tianjin, China). According to the regulation of XFZYD, the decoction pieces (total weight of 78 g) were mixed and immersed in 600 mL deionized water for 1 h at room temperature, and then refluxed twice for two hours per time. After filtration and concentration, aqueous extract was dried at 45 °C in an oven under vacuum to give 30 g extract powder, the yield of which was 38.5%. The extract powder was stored at 4 °C before use. Male Sprague-Dawley (SD) rats (200 ± 10 g) were supplied by Beijing HFK Bioscience Co., Ltd. (SCXK2009-0004, Beijing, China). All animals were acclimated in a room (22–25 °C with 50% ± 10% humidity) for one week, which had free access to water and standard laboratory food, and gradually adapted to the facilities. Then, the rats were fasted with free access to water for 12 h prior to the experiment. For seven consecutive days, the aqueous extract of XFZYD was administrated to the rats intragastrically at a dose of 16.38 g/kg body weight, which was expressed as the weight of decoction pieces ( Zhang et al., 2018 , Fan et al., 2020 ). After the last administration, the blood samples were respectively collected from the retroorbital venous plexuses at 0.5, 1 and 2 h, which were placed at room temperature for 1 h and centrifuged at 8000 rpm for 10 min. Then, the serum samples were obtained from the different blood sampling time in one rat and equally mixed to acquire the blended serum samples, which were frozen immediately and stored at −80 °C until analysis. Blank serum samples were collected in the same way. Animal studies were conducted according to protocols approved by the Review Committee of Animal Care and Use. The serum samples were thawed and homogenized at room temperature in advance. After addition of 10% formic acid (10 μL) in serum (1 mL), the sample was vortex-mixed well and loaded on Cleanert S C18-SPE column (500 mg/3 mL, Bonna-Agela, China), which was respectively pretreated with 2 mL methanol and 2 mL water, and equilibrated with 3 mL 1% formic acid in sequence. The loaded sample was sequently eluted with 3 mL 1% formic acid and 2 mL methanol, then methanol eluant was collected and evaporated to dryness by a vacuum centrifugal concentrator (Eppendorf, Hamburg, Germany) at 40 °C. Finally, the residue was dissolved in 50% methanol (100 μL) for analysis. All the MS data were acquired and processed by Masslynx V4.1 software (Waters Corporation, Milford, MA, USA). The structures of prototype compounds were drawn using ChemBioDraw 14 software (CambridgeSoft Corporation, USA). The network was portrayed by Cytoscape V3.8.0 software (National Resource for Network Biology, USA).

In this study, we have established a rapid and effective UPLC-Q-TOF/MS method to identify the prototype compounds and their metabolites in rat' serum after intragastric administration of XFZYD, from which 175 compounds (24 prototype compounds and 151 metabolites) were identified and tentatively characterized. Metabolic rules of compounds with the different type of structures were summarized from the relationship between the prototype compounds and their metabolites. This study is meaningful for clarification of the absorbed compounds in vivo . Also, this study has demonstrated the feasibility of UPLC-Q-TOF/MS application in exploring the prototype compounds and metabolites originated from compound formula.

Xuefu Zhuyu Decoction (XFZYD), developed by Qingren Wang (1768–1831) in Yilin Gaicuo (Correction of Medical Errors, 1850), is a well-known traditional Chinese medicine formula. XFZYD consists of eleven herbs, including Persicae Semen (Taoren), Angelicae Sinensis Radix (Danggui), Chuanxiong Rhizoma (Chuanxiong), Carthami Flos (Honghua), Paeoniae Radix Rubra (Chishao), Rehmanniae Radix (Dihuang), Aurantii Fructus (Zhiqiao), Bupleuri Radix (Chaihu), Platycodonis Rad ix (Jiegeng), Achyranthis Bidentatae Radix (Niuxi), and Glycyrrhizae Radix et Rhizoma (Gancao), which is widely used to treat diseases caused by qi stagnation and blood stasis with its effects of blood-activating, stasis-resolving and regulating qi -flowing ( Wang et al., 2022 , Wang, 2005 ). It was reported that XFZYD has the significant clinical effects on dysmenorrhoea, traumatic brain injury, and endometriosis ( Jo et al., 2017 , Zhang et al., 2018 , Fu et al., 2019 , Su et al., 2020 ). Especially, XFZYD is widely employed to treat cardiovascular diseases, including chest pain, headache, angina pectoris, heart failure, atherosclerosis, hypertension, and hyperlipidemia as the basic prescription ( Yang et al., 2019 , Lin et al., 2018 , Jiang and Jiang, 2016 , Wang et al., 2015 , Wang and Qiu, 2019 , Meng et al., 2018 ). However, its underlying action mechanism is still uncovered. Importantly, the clarification of chemical compounds in vivo is the prerequisite question which must be answered and worthy of in-depth study. In recent years, researchers have made continuous efforts to develop various analytical methods to clarify the chemical compounds of XFZYD. Previously, our group employed ultra performance liquid chromatography with diode array detector tandem mass spectrometry (UPLC-DAD-MS/MS) method to identify 28 compounds and quantitatively analyze 12 compounds in XFZYD related products ( Zhang et al., 2012 ). Also, we made use of ultra performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF/MS) to systematically illuminate the chemical compounds of XFZYD in vitro , by which 103 compounds were identified, mainly including phenolic acids, flavonoids, saponins, terpenes, and other compounds ( Zhang et al., 2015 ). Besides this, Fu et al (2016) qualitatively analyzed 34 major constituents, including organic acids, lactones, alkaloids, amino acids and cyanogenic glycosides in XFZYD by using ultra high performance liquid chromatography with hybrid ion trap time-of-flight mass spectrometry (UHPLC-ESI-IT-TOF-MS). However, there are few reports on clarification of the prototype compounds and their metabolites from XFZYD exposed in vivo , which will be important for illuminating the potentially active compounds. The metabolites always act as the active compounds in vivo. For instance, dihydroberberine, jatrorrhizine, columbamine, berberrubine, and demethyleneberberine, the metabolites of berberine produced by intestinal microbiota, have the significant effect of lipid-lowering ( Feng et al., 2018 , Zuo et al., 2006 , Zhou et al., 2014 ). Therefore, it is essential to study the exposed compounds from XFZYD in vivo . In this study, we employed UPLC-Q-TOF/MS for identification of chemical compounds and their metabolites in rats’ serum after oral administration of XFZYD aqueous extract at a dose of 16.38 g/kg body weight, which was expressed as the weight of decoction pieces. A total of 175 compounds were characterized and tentatively identified, including 24 prototype compounds and 151 metabolites. Also, the major metabolic pathways of prototype compounds from XFZYD in serum have been preliminarily proposed. Meaningfully, we established an analytical method with high sensitivity, rapid analysis, low consumption of samples, and abundant yield of structural information. Our study not only provides the information about the chemical substances of XFZYD in vivo , but also highly contributes to further investigation of the pharmacology and mechanism of XFZYD.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.