Translate this page into:
An integrated strategy for comprehensive characterization of chemical components in Qingqiao Kangdu granules by UHPLC-Q-Exactive-MS coupled with feature-based molecular networking
⁎Corresponding authors. yanfang303@163.com (Fang Yan), 20120941161@bucm.edu.cn (Wei Cai)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Qingqiao Kangdu granule (QQKDG), a traditional Chinese medicine (TCM), has been used clinically to treat various viral diseases, including flu, mumps, and viral hepatitis, owing to its abundant bioactivities. Nevertheless, the chemical components of QQKDG have not been sufficiently elucidated; consequently, the development of standards for quality evaluation and complete understanding of the pharmacological mechanisms of action are hindered. Therefore, a systematic approach must be developed to efficiently discover novel compounds and advance pharmacological research. In this regard, this study proposed an integrated strategy for the comprehensive characterization of the chemical components in QQKDG by UHPLC-Q-Exactive-MS coupled with feature-based molecular networking (FBMN) to improve annotation accuracy and achieve visualization. First, the chromatographic and mass spectrum conditions were optimized to obtain good separation and abundant signal response. Subsequently, an in-house library was established by searching for relevant literature to improve annotation confidence. Finally, the raw data acquired under optimized conditions were uploaded to the FBMN to achieve component visualization by connecting precursor ions of the same color, in which compounds have similar structural features. Thus, a total of 231 compounds, including 89 flavonoids, 36 phenolic acids, 26 phenylethanoid glycosides, 23 coumarins, 17 chlorogenic acid derivatives, 14 terpenoids, 10 alkaloids, 10 lignans and 6 other compounds, were characterized, and numerous novel compounds with new structures were explored. Thus, this study provides a strategy for comprehensive characterization, which can also be applied to other TCMs.
Keywords
Qingqiao Kangdu granule
Characterization
UHPLC-Q-Exactive-MS
Feature-based molecular networking
1 Introduction
Currently, increasingly number of complex diseases that are difficult to control and cure using the conventional drug development philosophy of single-compound–single-target, whereby a drug with only one compound targets a specific protein to treat a particular disease, are surfacing (Hu and Sun, 2017; Wang et al., 2014). Since centuries, traditional Chinese medicines (TCMs) have been used extensively in China and other Asian countries to treat difficult miscellaneous diseases; they comprise several medicinal herbs mixed in a specific mass ratio according to the rules of the monarch, minister, assistant, and guide (Pang et al., 2016; Wang et al., 2018). Recently, TCMs have received considerable attention in treating complicated and chronic diseases owing to their multicomponent, multipathway, and multitarget effects (Wang et al., 2020). For example, Yupingfeng San is a TCM formula that consists of three herbs, Huangqi, Fangfeng, and Baizhu in Chinese, which improve lung Qi, relieve pain, and enhance the spleen, owing to the antiimmunity, antiinflammatory, and gastrointestinal tract regulation effects based on the chemical components of flavonoids, chromones, and sesquiterpenoids in lung diseases (Zhang et al., 2015; Aravilli et al., 2017; Xu and Zhang, 2020; Yang et al., 2021). Several TCMs can be combined with Western medicines to relieve clinical symptoms and reduce toxicity and side effects, thereby improving the quality of life of patients (Xu and Chen, 2010).
Qingqiaokangdu granule (QQKDG), a TCM formula, was developed by the hospital affiliated with the Chengdu University of TCM according to the local climatic characteristics and features of resident crowd. This QQKDG is composed of fourteen herbs, namely Lonicera japoraca (yinhuateng), Forsythia suspensa (lianqiao), Pueraria montana (fenge), Angelicae dahuricae (baizhi), Artemisia caruifolia (qinghao), Radix Bupleuri (chaihu), Paridis Rhizoma (chonglou), the dried root of Isatis tinctoria (banlangen), the dried rhizoma of Iris tectorum (chunshegan), Taraxacum mongolicum (pugongying), the dried leaves of Isatis indigotica (daqingye), Pogostemon cablin (guanghuoxiang), Perillae Folium (zisuye), and Mentha canadensis (bohe) (Xiong et al., 2014). Modern pharmacological research has demonstrated that QQKDG has the effects of clearing heat, detoxifying, relieving external heat, and reducing fever; thus, it has been extensively used in treating viral diseases, including flu, mumps, and viral hepatitis (Xia et al., 2016). Its efficacy and safety have been proven in double-blind randomized, controlled clinical trials (Xiong, 2015). Although QQKDG has remarkable efficacy on anemopyretic cold, it contains numerous unknown chemical compounds, owing to which, elucidating the therapeutic material basis and action mechanisms, and their globalization are both challenging. Xia et al. developed a method based on high-performance liquid chromatography (HPLC) switching wavelengths to simultaneously determine the content of seven constituents in QQKDG (Xia et al., 2016). Therefore, the chemical compounds in QQKDG must be determined, which can be useful for quality control and standardization.
Ultra-high-performance liquid chromatography (UHPLC) coupled with high-resolution mass spectrometry (HRMS) is a powerful analytical tool for identifying and characterizing the chemical compounds in TCMs owing to its strong separation ability and structure prediction (Fu et al., 2021; Gao et al., 2021). Computer-aided software, developed primarily for targeted screening based on known databases, has the advantage of complex data processing and has greatly promoted the characterization of chemical components in TCMs, such as waters UNIFI and thermo fisher compound discoverer (Chen et al., 2021). Global natural products social (GNPS, https://gnps.ucsd.edu/) molecular networking (MN) is an open-access knowledge platform for the rapid clustering and analysis of mass spectrometry data; it was developed to comprehensively characterize the chemical ingredients of TCMs according to MS/MS spectrum similarity and public databases (Zhang et al., 2021). In addition, MN was applied to speculate unannotated nodes by relating them to structural analogs of annotated compounds based on MS data and fragmentation pathways (Zhang et al., 2022). Feature-based molecular networking (FBMN), a novel data analysis tool in MN, has been employed to distinguish isomers, essentially providing several advantages in visualizing and clustering unknown compounds (Li et al., 2022). It is beneficial for rapidly identifying compounds in complex systems and discovering unknown components owing to their strong integration and classification abilities. Therefore, in this study, an integrated strategy was established to rapidly detect and characterize the chemical components of QQKDG using UHPLC-Q-Exactive-MS coupled with FBMN. This is the first systematic investigation of the chemical composition of QQKDG, and the results provide a comprehensive understanding of the material basis of QQKDG against anemopyretic colds. This strategy also provides an efficient method for identifying ingredients in TCM prescriptions.
2 Materials and methods
2.1 Materials and reagents
QQKDG were obtained from the Hospital of the Chengdu University of TCM (Chengdu, China; batch number: 20220210). A total of 42 reference standards, including 22 flavonoids, 7 phenylpropanoids, 5 organic acids, 3 coumarins, 1 lignan, 1 terpenoid, 1 alkaloid, 1 paraben, and 1 phenylethanoid glycoside, were characterized by 1HNMR, 13C NMR, and MS spectral analyses, and the purities were above 98 % by HPLC analysis. Detailed information regarding the reference standards is provided in Supplementary Table S1. Chromatography-grade methanol and acetonitrile were purchased from Merck (Kenilworth, NJ, USA). Distilled water was obtained from Guangzhou Watson Food and Beverage Co., Ltd. (Guangzhou, China). LC-MS-grade formic acid was purchased from Fisher Scientific (Waltham, MA). All the other reagents were of analytical grade.
2.2 Sample preparation
QQKDG (5 g) was ultrasonically extracted with 70 % methanol (100 mL) for 1 h at room temperature. The extracting solution was concentrated at 50 °C using a rotary vacuum evaporator and water was removed using a freeze dryer to obtain the residue of QQKDG. Subsequently, the portion of QQKDG was dissolved in methanol and centrifuged at 12000 rpm for 20 min, and filtered through a 0.22-μm millipore filter before LC-MS analysis.
The 42 reference standards were dissolved in methanol at an approximate concentration of 1 mg/mL. Each stock solution was mixed and diluted in methanol to obtain the standard mixture solution (approximately 20 μg/mL). Subsequently, the mixed solution was centrifuged at 12000 rpm for 20 min, and the supernatant was stored at 4 °C before analysis.
2.3 Liquid chromatographic conditions
Chromatographic separation of the sample was performed using an Ultimate 3000 UHPLC system (Thermo Fisher Scientific, California, USA) equipped with a binary pump, autosampler, degasser, and column compartment. The separation of the QQKDG was performed on a Thermo Scientific Syncronis C18 (100 mm × 2.1 mm, 1.7 μm) at 45 °C. The mobile phase consisted of 0.1 % formic acid in water (A) and acetonitrile (B). The flow rate was 0.28 mL/min, and the gradient elution program was optimized as: 0–2 min, 5–10 % B; 2–5 min, 10–15 % B; 5–10 min, 15–20 % B; 10–12 min, 20–40 % B; 12–20 min, 40–55 % B; 20–25 min, 55–80 % B; 25–26 min, 80–5 % B; and 26–30 min, 5 % B. The injection volume was 2 µL.
2.4 MS spectrometry conditions
Mass spectrometry was performed using a Q-Exactive Orbitrap MS (Thermo Fisher Scientific, Bremen, Germany) instrument equipped with a heated electrospray ionization source (HESI). Data acquisition progressed in both positive and negative ion modes through full-scan data-dependent MS/MS (full scan-ddMS2) with a mass range of m/z 100–1500. Mass spectrometry conditions were set as follows. The capillarycc voltage was set as 3.5 kV in the positive ion mode and 3.0 kV in the negative ion mode; the full mass resolution was set to 70000; the heater temperature and heated capillary temperature were 350 and 320 °C, respectively; the sheath gas and auxiliary gas flow rate were 30 and 10 arb, respectively; the S-lens radio frequency (RF) level was 50; and the ddMS2 resolution was set to 17500. The stepped normalized collision energies (NCEs) of 30, 40, and 60 % were employed for fragmentation. Xcalibur 4.2 software (Thermo Fisher Scientific, California, USA) was used for data acquisition and analysis.
2.5 Feature-based molecular networking analysis
The UHPLC-Q-Exactive-MS raw data in positive and negative ionization modes were converted to mzXML format with MS conversion and then imported into MZmine (version 2.53) for chromatographic feature extraction. The MZmine filter parameters are listed in Supplementary Table S2. Two files, the feature quantification table (.CSV file) with peak areas and MS2 spectral summaries (.MGF file) with a representative MS2 spectrum were exported from MZmine; subsequently, the MS2 file MGF, feature quantification table, and original mzML were imported into the GNPS FBMN analysis website (https://gnps-quickstart.ucsd.edu/featurebasednetworking) and visualized using Cytoscape 3.9.1. The molecular networking parameters included a precursor ion mass tolerance of 0.02 Da, minimum matched peaks > 6, cosine score of > 0.7, and library search matched peaks > 3.
3 Results and discussion
3.1 Integrated strategy for data analysis
QQKDG is composed of 14 herbs containing tens of thousands of compounds. Thus, herein, an effective strategy was established to comprehensively and accurately characterize the chemical constituents of QQKDG in Fig. 1. This strategy comprises three steps. First, to achieve good separation and abundant signal response, the proportion and variety of mobile phases, including acetonitrile-aqueous, methanol-aqueous, acetonitrile-aqueous with 0.1 % formic acid, and methanol-aqueous with 0.1 % formic acid, were optimized to obtain better chromatographic conditions. Thus, the mobile phases consisted of acetonitrile-aqueous with 0.1 % formic acid, which could be considered as the most optimized separation condition, and the proportion was in “2.3 Liquid chromatographic conditions.” Second, information regarding the compound names, molecular formulae, and exact MS/MS fragment ions of the chemical components derived from the 14 herbs of QQKDG were obtained from PubChem (https://pubchem.ncbi.nlm.nih.gov), CNKI (https://www.cnki.net), Web of Science (https://www.webofknowledge.com), and Google Scholar (https://scholar.google.com/) to develop an in-house library of QQKDG. The unknown compounds were identified by comparing the quasi-molecular ion and MS2 fragment ions using an in-house library, and the fragmentation patterns were summarized based on existing standards. Third, GNPS were used to detect and identify unknown compounds based on fragment similarity. Raw data, including MS and MS/MS spectra, were converted into mzXML format and uploaded to mzmine software for data preprocessing; subsequently, the files of the quantification table and MS2 spectral summary were imported into the GNPS-FBMN platform. In addition, the MS2 fragment ions were matched with the GNPS database for a fast and accurate analysis. Unannotated compounds were identified according to the MS/MS spectral relevance between the annotated compounds and in-house library.
An integrated strategy for chemical characterization of QQKDG extract.
3.2 Characterization of chemical constituents in QQKDG by UHPLC-Q-Exactive-MS based on FBMN
In this study, 231 compounds were detected, of which 42 compounds were precisely identified and 189 compounds were putatively identified using the integrated strategy; these compounds included 89 flavonoids, 36 phenolic acids, 25 phenylethanoid glycosides, 23 coumarins, 17 chlorogenic acid derivatives, 14 terpenoids, 10 alkaloids, 10 lignans, and 7 others. Detailed information of components, including peak number, retention time (Rt), accurate molecular ions, formulas, mass errors (within ± 5 ppm), fragment ions and compound names, are presented in Table 1 and Supplementary Table S3. High-resolution extracted ion chromatograms (HREICs) of QQKDG in both positive and negative ion modes are shown in Fig. 2. In addition, the raw data in the two ion modes were processed using FBMN, an advanced processing method for distinguishing isomers with similar MS2 spectra (Qu et al., 2023). Compounds of the same type could be clustered together and annotated by matching the MS/MS fragment ions of the FBMN database. Therefore, a comprehensive FBMN of QQKDG based on MS/MS spectral similarity was obtained, as shown in Supplementary Fig. S1. The molecular map contained a total of 2867 precursor ions, including 354 clusters (nodes, ≥2) and 3580 edges in negative mode, and a total of 2936 precursor ions, including 298 clusters (nodes, ≥2) and 4282 edges in positive mode. More detailed information is available on the open website (https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=445f9d16f4774ce0a7be99730bc6dcf1 in the negative mode; https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=c4b076f753b944dd94b8101de8abafe0 in the positive mode).
peak
tR
Theoretical Mass m/z
Experimental Mass m/z
Error (ppm)
Formula
Identification
peak
tR
Theoretical Mass m/z
Experimental Mass m/z
Error (ppm)
Formula
Identification
1*
0.86
191.05611
191.05501
−5.76
C7H12O6
Quinic acid
117*
10.80
623.19814
623.19745
−1.11
C29H36O15
Forsythoside A
2
0.92
191.01972
191.01875
−5.11
C6H8O7
Isocitric acid
118*
10.88
593.15119
595.16574593.15088
595.16370−0.53
−3.43C27H30O15
Kaempferol-3-O-neohesperidoside
3*
1.32
191.01972
191.01874
−5.16
C6H8O7
Citric acid
119*
10.95
623.19814
623.19751
−1.02
C29H36O15
Acteoside
4
1.35
344.04015
344.03983
−0.95
C10H12N5O7P
cGMP
120
10.97
461.14422
461.14096
−7.08
C23H24O10
Kuzubutenolide A
5
1.76
361.11402
361.11368
−0.94
C15H22O10
3,4-DihydroxylPhenyle-thanoidglycoside
121
11.04
417.11800
417.11646
−3.71
C21H20O9
Neopuerarin
6*
2.07
169.01424
169.01321
−6.13
C7H6O5
Gallic acid
122*
11.04
463.08819465.10275
463.08792465.10114
−0.60
−3.46C21H20O12
Isoquercitrin
7
2.32
335.13475
335.13464
−0.35
C14H24O9
Rengynic acid-1′-O-β-D-glucoside
123
11.14
461.07254
461.07239
−0.35
C21H18O12
Scutellarin
8
2.80
315.07215
315.10834
−0.64
C13H16O9
protoeatechuic acid-O-glucoside isomer
124
11.25
461.07254
461.07242
−0.28
C21H18O12
Luteolin 7-O-glucuronide
9
2.85
331.06706
331.06699
−0.24
C13H16O10
Gallic acid-4-O-glucoside
125
11.28
623.16175
625.17631623.15979
625.17401−3.15–3.68
C28H32O16
Isorhamnetin-3-O-neohespeidoside
10
3.29
315.07215
315.07199
−0.52
C13H16O9
protoeatechuic acid-O-glucoside isomer
126
11.31
447.09328449.10783
447.09308449.10645
−0.46
−3.09C21H20O11
Luteolin-5-O-glucoside
11
3.33
197.04554
197.04460
−4.80
C9H10O5
Danshensu
127
11.43
247.06009
247.05942
−2.75
C13H10O5
Pimpinellin
12
3.38
515.14063
517.15518515.14008
517.15381−1.07–2.65
C22H28O14
caffeoylquinic acid-hexoside
128
11.51
477.10384
477.10367
−0.38
C22H22O12
Genistein-7-O-glucoside
13
3.53
329.08780
329.08765
−0.47
C14H18O9
vanillic acid -O-glucopyranoside
129*
11.52
431.09837433.11292
431.09778433.11130
−1.37–3.74
C21H20O10
Apigenin-7-O-β-D-glucoside
14
3.85
315.07215
315.07202
−0.43
C13H16O9
protoeatechuic acid-O-glucoside isomer
130
11.55
769.25605
769.25568
−0.48
C35H46O19
Forsythoside G
15
4.06
375.12967
375.12912
−1.47
C16H24O10
8-Epiloganic acid
131
11.57
653.17232
655.18687653.17181
655.18469−0.79–3.33
C29H34O17
Iristectorigenin B-7-O-β-glucosyl (1 → 6) glucoside isomer
16
4.11
515.14063
517.15518515.14001
517.15350−1.20–3.25
C22H28O14
caffeoylquinic acid-hexoside
132
11.58
477.14023
477.13931
−1.94
C23H26O11
Calceolarioside B
17
4.16
505.15627
505.15610
−0.35
C21H30O14
Hebitol II
133
11.61
247.09648
247.09560
−3.58
C14H14O4
Marmesin
18
4.31
311.04085
311.04062
−0.76
C13H12O9
Caftaric acid
134
11.88
607.20322
607.20343
0.33
C29H36O14
Lipedpside A
19*
4.33
353.08780355.10235
353.08740355.10101
−1.15–3.79
C16H18O9
Neochlorogenic acid
135
11.92
453.14023
[M−H + HCOOH]-453.14005
−0.41
C20H24O9
Nodakenin
20
4.43
447.15079
447.15048
−0.71
C19H28O12
rebouoside B
136
11.93
247.09648
247.09558
−3.66
C14H14O4
Decursinol
21
4.49
341.08781
341.08768
−0.37
C15H18O9
caffeic acid-O-glucoside
137
12.02
445.11402
445.11356
−1.03
C22H22O10
3′-Methoxy puerarin
22
4.60
515.14063
517.15518515.14014
517.15356−0.95–3.13
C22H28O14
caffeoylquinic acid-hexoside
138
12.05
515.11949
517.13405515.11859
−1.77
C25H24O12
1, 4-Dicaffeoylquinic acid
23
4.74
299.11362
299.11334
−0.96
C14H20O7
Salidroside
139*
12.11
461.10893463.12348
461.10834463.12137
−1.29
−4.57C22H22O11
Tectoridin
24
4.82
461.16644
461.16586
−1.28
C20H30O12
Forsythoside E
140
12.22
477.10384
479.11840477.10379
479.11661−0.12
−3.74C22H22O12
Isorhamnetin-5-O-glucoside
25
4.91
339.07215
339.07156
−1.76
C15H16O9
Esculin
141
12.31
503.11840
503.11658
−3.62
C24H22O12
6′′-O-Malonyl daidzin
26
4.92
327.13393
327.13235
−8.19
C18H18N2O4
isaindigodione isomer
142*
12.32
593.15119
595.16574593.15070
595.16364−0.83
−3.53C27H30O15
Kaempferol-3-O-rutinoside
27
4.92
535.16684
[M−H + HCOOH]-535.16620
−1.20
C21H30O13
Tectoruside
143
12.36
507.11441
[M−H + HCOOH]-507.11386
−1.09
C22H22O11
Isotectorigenin-O-glucoside
28
4.93
579.17083
579.16840
−4.20
C27H30O14
Puerarin-4′-O-glucoside
144*
12.39
515.11949517.13405
515.11884517.13226
−1.28–3.46
C25H24O12
Isochlorogenic acid B
29
4.95
375.12967
375.12906
−1.63
C16H24O10
Loganic acid
145*
12.52
515.11949517.13405
515.11884517.13239
−1.28–3.21
C25H24O12
Isochlorogenic acid A
30
5.21
407.09837
407.09793
−1.30
C19H20O10
Iriflophenone-2-O-β-glucoside
146
12.54
623.19814
623.19757
−0.92
C29H36O15
Forsythoside I
31
5.33
375.12967
375.12924
−1.15
C16H24O10
Adoxosidic acid
147
12.55
357.13436
357.13370
−1.85
C20H22O6
Epipinoresinol
32
5.40
355.10345
355.10336
−0.27
C16H20O9
ferulic acid-O-glucoside
148
12.55
519.18718
519.18646
−1.40
C26H32O11
Pinoresinol-O-glucoside
33
5.44
337.09289
337.09280
−0.27
C16H18O8
p-coumaroylquinic acid
149
12.56
477.10384
479.11840477.10342
−0.90
C22H22O12
Isorhamnetin-3-O-glucoside
34
5.48
709.19853
709.19836
−0.25
C32H38O18
Mirificin-4′-O-glucoside
150
12.56
538.22828
[M + NH4]+538.22638
−3.54
C26H32O11
Matairesinol-4-O-D-glucopyranoside
35
5.51
341.08781
341.08771
−0.28
C15H18O9
caffeic acid-O-glucoside
151
12.57
247.09648
247.09549
−4.03
C14H14O4
Columbianetin
36
5.60
295.04594
295.04562
−1.09
C13H12O8
coumaroyl-tartaric acid
152
12.60
607.20322
607.20325
0.04
C29H36O14
Forsythenside K
37
5.62
389.10893
389.10855
−0.99
C16H22O11
secologanoside
153
12.68
623.16175
623.16022
−2.47
C28H32O16
Isorhamnetin-3-O-β-D-rutinoside
38
5.74
327.13393
327.13245
−7.89
C18H18N2O4
isaindigodione isomer
154
12.72
433.11402
433.11383
−0.44
C21H22O10
Naringenin-7-O-glucoside
39
5.74
579.17083
579.16858
−3.89
C27H30O14
Daidzein-4,7-O-glucoside
155
12.74
491.11949
491.11890
−1.22
C23H24O12
Iristectorin B
40*
5.79
353.08780355.10235
353.08734
−1.32
C16H18O9
Chlorogenic acid
156
12.81
349.08636
349.08618
−0.53
C16H18N2O5S
Indole-3-acetonitrile-2-S-β-D-glucopyranoside
41
5.82
373.11402
[M−H + HCOOH]-373.11349
−1.42
C15H20O8
Androsin
157
12.90
503.17701
[M−H + CH3COOH]-503.17767
1.30
C20H28O11
Hyuganoside Ⅳ
42
5.98
253.07176
253.07141
−1.39
C12H14O6
hwanggeumchal B isomer 1
158
12.95
477.10384
479.11840477.10330
−1.15
C22H22O12
Isorhamnetin-7-O-glucoside
43*
6.02
353.08780355.10235
353.08737355.10068
−1.23–4.72
C16H18O9
Cryptochlorogenic acid
159
12.95
507.11441
[M−H + HCOOH]-507.11389
−1.03
C22H22O11
Isotectorigenin-7-O-β-D-glucoside isomer 2
44
6.03
355.10235
355.10068
−4.73
C16H18O9
Scopolin
160*
12.98
445.07763447.09218
445.07724447.09067
−0.89
−3.39C21H18O11
Apigenin-7-O-glucuronide
45
6.03
507.17192
507.17133
−1.18
C21H32O14
secologanoside A
161
12.99
717.14610
717.14618
0.10
C36H30O16
Salvianolic acid B
46
6.05
373.11402
373.11356
−1.23
C16H22O10
Swertiamarin
162*
13.04
515.11949517.13405
515.11890517.13239
−1.16–3.21
C25H24O12
Isochlorogenic acid C
47
6.26
367.10345
369.11800367.10318
369.11664−0.75
−3.70C17H20O9
Feruloylquinic acid
163
13.10
719.16175
719.16113
−0.87
C36H32O16
Sagerinic acid
48
6.40
177.01933
177.01837
−5.43
C9H6O4
Daphnetin
164*
13.10
359.07724
359.07681
−1.20
C18H16O8
Rosmarinic acid
49*
6.50
179.03498
179.03394
−5.82
C9H8O4
Caffeic acid
165
13.14
491.11949
491.11890
−1.22
C23H24O12
Iristectorin A
50
6.67
355.10345
355.10355
0.27
C16H20O9
ferulic acid-O-glucoside
166
13.20
521.13006523.14461
521.12952523.14252
−1.04–4.00
C24H26O13
Iridin
51
6.73
593.15119
595.16574593.15100
595.16351−0.33–3.75
C27H30O15
Glucosylvitexin
167
13.21
267.07641
267.07529
−4.23
C15H10O3N2
3-(2′-carboxyPhenyl)-4-(3H)-quinazolinone
52
6.79
627.15557
627.15350
−3.31
C27H30O17
168*
13.21
435.12967
435.12897
−1.61
C21H24O10
Phlorizin
53
6.84
579.17083
579.16876
−3.58
C27H30O14
6″-O-α-D-glucopyranosylpuerarin
169
13.24
459.12857
459.12683
−3.80
C23H22O10
6″-O-Acetyl daidzin
54
6.89
481.09876
481.09842
−0.72
C21H22O13
gallic acid-3-methyl ether-4-O-protocatechuoylglucoside
170
13.26
621.18249
621.18250
0.01
C29H34O15
Suspensaside A
55
6.92
579.17083
579.16852
−3.99
C27H30O14
3′-Methoxypuerarin-O-apioside
171
13.34
519.18718
519.18701
−0.34
C26H32O11
Matairesinoside
56*
6.95
342.16998
[M]+342.16867
−3.84
C20H24O4N
Magnoflorine
172
13.39
473.14532
473.14520
−0.25
C24H26O10
Sophoraside A
57
6.96
253.07176
253.07141
−1.39
C12H14O6
hwanggeumchal B isomer 2
173
13.39
519.15079
[M−H + HCOOH]-519.15100
0.39
C25H28O12
Pueroside C
58*
7.06
417.11800
415.10345417.11606
415.10291−4.67
−1.31C21H20O9
Puerarin
174
13.40
373.12927
373.12967
1.06
C20H22O7
(+)-1-Hydroxylpinoresinol
59
7.12
639.19305
639.19287
−0.29
C29H36O16
R-suspensaside
175
13.61
315.05102
315.05084
−0.59
C16H12O7
Irilin D
60
7.13
707.25215
[M + Na]+707.24933
−4.00
C32H44O16
Lariciresinol-４,４′ -O-β-D-diglucoside
176
13.62
431.13365
431.13217
−3.45
C22H22O9
Ononin
61
7.14
729.26113
[M−H + HCOOH]-729.26111
−0.04
C32H44O16
Clemastanin B
177
13.68
467.21339
467.21368
0.60
C20H36O12
N-octanoylsucrose
62
7.16
387.16605
387.16574
−0.82
C18H28O9
Jasmonic acid-5′-O-glucoside
178
13.81
239.08150
239.08054
−4.03
C14H10O2N2
3-(2′-Hydroxypheny)-4-(3H)-quinazolinone
63
7.28
639.19305
639.19318
0.19
C29H36O16
isocampneoside II
179
13.83
431.13365
431.13223
−3.31
C22H22O9
Isoononin
64*
7.33
403.12458
[M−H + HCOOH]-403.12405
−1.33
C16H22O9
Sweroside
180
13.87
283.15509
283.15482
−0.98
C15H24O5
Dihydroartemisinin
65
7.35
435.15079
[M−H + HCOOH]-435.15027
1.30
C17H26O10
Loganin
181
13.90
255.06518
255.06404
−4.49
C15H10O4
Daidzein
66
7.39
337.09289
339.10744337.09283
339.10574−0.18
−5.02C16H18O8
p-Coumaroylquinic acid
182*
13.90
579.20831
[M−H + HCOOH]-579.20770
−1.06
C27H34O11
Forsythin
67
7.46
593.15119
595.16574593.15100
595.16364−0.33–3.53
C27H30O15
Vicenin II
183
13.96
203.03388
203.03311
−3.82
C11H6O4
Xanthotoxol
68
7.51
639.19305
639.19293
−0.20
C29H36O16
S-suspensaside
184
13.96
287.09140
287.09018
−4.25
C16H14O5
Oxypeucedanin
69*
7.56
547.14571
549.16026547.14539
549.15802−0.59–4.09
C26H28O13
Puerarin apioside
185
13.96
305.10196
305.10077
−3.92
C16H16O6
Prangenin hydrate
70*
7.63
515.11949517.13405
515.11902517.13257
−0.93–2.86
C25H24O12
1,3-Dicaffeoylquinic acid
186
13.96
504.18751
504.18851
1.97
C25H31NO10
L-Phenylalaninosecologanin B
71
7.67
653.17232
655.18687653.17212
655.18488−0.31–3.04
C29H34O17
Iristectorin B-4′-O-glucoside
187
14.03
677.15119
677.15070
−0.73
C34H30O15
3, 4, 5-O-tricaffeoylquinic acid
72
8.01
637.10463
637.10455
−0.14
C27H26O18
Scutellarein-7-O-diglucuronide
188
14.08
319.11761
319.11630
−4.12
C17H18O6
3′-O-Acetylhamaudol
73
8.09
639.19305
639.19312
0.10
C29H36O16
Lugrandoside
189
14.09
653.17232
655.18687653.17151
655.18469−1.24
−3.33C29H34O17
Iristectorigenin B-7-O-β-glucosyl(1 → 6)-glucoside isomer
74
8.30
239.09249
239.09195
−2.29
C12H16O5
3,4′-Dihydroxy-3′-methoxy-benzenepentanoic acid
190*
14.20
285.04046
285.04013
−1.16
C15H10O6
Luteolin
75
8.30
579.17083
579.16882
−3.47
C27H30O14
3′-Methoxydaidzin-O-apioside
191*
14.21
301.03537
301.03500
−1.25
C15H10O7
Quercetin
76
8.31
593.15119
595.16574593.15094
595.16345−0.43–3.85
C27H30O15
Glucosyl-vitexin
192
14.36
305.10196
305.10083
−3.72
C16H16O6
Oxypeucedan hydrate
77
8.31
639.19305
639.19263
−0.67
C29H36O16
R-campneoside II
193
14.54
207.06628
207.05037
−3.17
C11H12O4
Ethyl caffeate
78
8.37
403.12458
403.12405
−1.33
C17H24O11
Secoxyloganin
194
14.57
335.11252
335.11087
−4.95
C17H18O7
Byakangelicin
79
8.37
563.14062
565.15518563.14020
565.15283−0.76
−4.16C26H28O14
Schaftoside
195*
14.57
317.10196
317.10046
−4.75
C17H16O6
Byakangelicol
80
8.37
637.10463
637.10419
−0.70
C27H26O18
Luteolin-7-O-diglucuronide
196*
14.93
271.06119
271.06100
−0.73
C15H12O5
Naringenin
81
8.39
367.10345
369.11800367.10297
369.11646−1.32–4.19
C17H20O9
Feruloylquinic acid
197*
15.02
269.04554
269.04532
−0.84
C15H10O5
Apigenin
82
8.45
639.19305
639.19348
0.66
C29H36O16
S-campneoside II
198
15.05
387.14492
387.14471
−0.56
C21H24O7
5′-O-caffeyl-jasmonic acid
83
8.66
449.14532
449.14478
−1.20
C22H26O10
Forsythenside F
199*
15.22
285.04046
285.04010
−1.27
C15H10O6
Kaempferol
84
8.69
417.11800
417.11612
−4.53
C21H20O9
Daidzin
200
15.23
299.05611
299.05566
−1.51
C16H12O6
Tectorigenin
85
8.78
639.19305
639.19275
−0.48
C29H36O16
Isolugrandoside
201*
15.50
315.05102
315.05084
−0.59
C16H12O7
Isorhamnetin
86
8.81
563.14062
565.15518563.14038
565.15302−0.44–3.82
C26H28O14
Vicenin III
202
15.56
329.06667
329.06635
−0.99
C17H14O7
Iristectorigenin A
87*
8.92
447.09328449.10783
447.09293449.10663
−0.79
−2.68C21H20O11
Orientin
203
15.62
277.10705
277.10587
−4.26
C15H16O5
3′(R)-+-Hamaudol
88
9.07
187.08658
187.08591
−3.63
C11H10N2O
Deoxyvasicinone
204
15.65
217.04953
217.04863
−4.17
C12H8O4
Bergapten
89
9.16
447.09328449.10783
447.09290449.10660
−0.86–2.75
C21H20O11
Homoorientin
205*
15.78
359.07724
359.07690
−0.95
C18H16O8
Irigenin
90
9.17
477.06746
477.06732
−0.30
C21H18O13
Quercetin-3-O-β-D-Glucuronide
206
15.79
329.06667
329.06647
−0.63
C17H14O7
Iristectorigenin B
91
9.26
535.18209
535.18195
−0.28
C26H32O12
(+)-Hydroxylpinoresinol-4′-O-glucopyranoside
207
16.06
359.07724
359.07700
−0.67
C18H16O8
Sudachitin
92
9.32
431.09837433.11292
431.09790433.11136
−1.09–3.60
C21H20O10
Vitexin
208
16.43
279.07641
279.07541
−3.62
C16H10N2O3
Hydroxy lindirubin
93
9.36
447.09328449.10783
447.09290449.10620
−0.86
−3.647C21H20O11
Kaempferol-7-O-β-D-glucopyranoside
209
16.44
267.06628
267.06580
−1.81
C16H12O4
Formononetin
94
9.38
609.18249
609.18225
−0.40
C28H34O15
Forsythoside J
210
16.60
217.04953
217.04887
−3.07
C12H8O4
Xanthotoxin
95
9.45
579.17192
579.17163
−0.52
C27H32O14
Naringin
211
16.68
247.06009
247.05922
−3.562
C13H10O5
Isopimpinellin
96
9.47
477.10384
479.11840477.10364
−0.44
C22H22O12
3′-Hydorxytectoridin
212
17.18
327.05102
327.05075
−0.84
C17H12O7
Iriflogenin
97
9.51
563.14062
565.15518563.14020
−0.76
C26H28O14
Isoschaftoside
213
17.36
249.06585
249.06487
−3.95
C15H8N2O2
Tryptanthrin
98
9.64
639.19305
639.19299
−0.11
C29H36O16
Plantamajoside isomer
214
17.39
825.46419
825.46454
0.42
C42H68O13
Saikosaponin A
99*
9.68
447.09328449.10783
447.09293449.10641
−0.79
−3.17C21H20O11
Luteolin-7-O-glucoside
215
17.93
233.04444
233.04349
−4.12
C12H8O5
5-Methoxy-8-hydroxypsoralen
100
9.69
607.20213
607.20044
−2.79
C29H34O14
Pueroside A
216
18.07
359.07614
359.07474
−3.91
C18H14O8
Dichotomitin
101
9.74
187.03897
187.03841
−3.00
C11H6O3
Isopsoralen
217*
18.30
191.10665
191.10577
−4.64
C12H14O2
Ligustilide
102
9.80
473.07254
473.07196
−1.25
C22H18O12
Cichoric acid
218
18.37
373.09289
373.09253
−0.97
C19H18O8
Junipegenin C
103
9.83
623.16175
625.17631623.16113
625.17365−1.01–4.25
C28H32O16
Tectorigenin-7-O-β-glucosyl (1 → 6) glucoside
219*
18.56
283.06119
283.06097
−0.80
C16H12O5
Oroxylin A
104
9.86
187.03897
187.03812
−4.55
C11H6O3
Psoralen
220
18.87
283.06119
283.06094
−0.91
C16H12O5
Genkwanin
105
9.91
609.18249
609.18231
−0.30
C28H34O15
Calceolarioside C
221
18.92
249.14852
249.14734
−4.74
C15H20O3
Arteannuin
106
10.02
477.14023
477.13956
−1.41
C23H26O11
Calceolarioside A
222
19.19
263.08150
263.08047
−3.93
C16H10N2O2
Indigo
107
10.04
193.04953
193.04874
−4.12
C10H8O4
Scopoletin
223
19.74
867.47475
867.47522
0.53
C44H70O14
AcetylSaikosaponin
108
10.05
623.19814
623.19769
−0.73
C29H36O15
Forsythoside H
224
19.91
359.11252
359.11115
−3.84
C19H18O7
5-Hydroxy-3′,4′,6,7-tetramethoxyFlavone
109
10.08
417.11800
417.11609
−4.60
C21H20O9
Puerarin isomer
225
20.12
263.08150
263.08035
−4.39
C16H10N2O2
Indirubin
110
10.17
609.18249
609.18243
−0.10
C28H34O15
Calceolarioside C
226
20.20
345.09687
345.09543
−4.20
C18H16O7
Penduletin
111
10.25
417.11800
417.11636
−3.95
C21H20O9
Daidzein 4′-O-glucoside
227
21.08
389.12309
389.12155
−3.97
C20H20O8
Artemetin
112
10.48
755.24040
755.23975
−0.86
C34H44O19
Forsythoside B
228*
21.38
271.09648
271.09537
−4.12
C16H14O4
Isoimperatorin
113*
10.49
609.14610
609.14575
−0.59
C27H30O16
Rutin
229
22.42
301.10705
301.10580
−4.15
C17H16O5
Cnidilin
114*
10.62
463.08819465.10275
463.08807465.10141
−0.28
−2.88C21H20O12
Hyperoside
230
23.04
359.11252
359.11090
−4.54
C19H18O7
Corymbosin
115
10.74
431.09837433.11292
431.09784433.11111
−1.23–4.18
C21H20O10
Isovitexin
231
23.28
223.09758
223.09694
−2.88
C12H16O4
Pogostone
116
10.80
223.06009
223.05899
−4.98
C11H10O5
Saikochromone A
The high resolution extracted ion chromatograms (HREICs) of QQKDG in the positive (P) and negative ion modes (N). N1 m/z 179.03498, 191.01972, 299.05611, 315.05102, 353.08780, 359.07724, 375.12967, 403.12458, 415.10345, 435.15079, 461.10893, 461.16644, 507.11441, 515.11949, 519.18718, 535.16684, 579.20831, 609.14610, 623.19814, 719.16175; N2: m/z 177.01933, 191.05611, 253.07176, 269.04554, 285.04046, 301.03537, 329.06667, 361.11402, 367.10345, 373.11402, 387.16605, 389.10893, 431.09837, 449.14532, 461.07254, 473.07254, 477.10384, 477.14023, 491.11949, 515.14063, 547.14571, 593.15119, 609.18249, 637.10463; N3: m/z 197.04554, 239.09249, 283.15509, 311.04085, 315.07215, 335.13475, 341.08781, 357.13436, 445.07763, 445.11402, 447.09328, 447.15079, 463.08819, 481.09876, 505.15627, 507.17192, 521.13006, 563.14062, 607.20322, 621.18249, 623.16175, 639.19305, 717.14610; N4: m/z 169.01424, 207.06628, 223.09758, 267.06628, 271.06119, 283.06119, 295.04594, 299.11362, 327.05102, 329.08780, 331.06706, 337.09289, 339.07215, 344.04015, 349.08636, 355.10345, 373.09289, 373.12927, 387.14492, 407.09837, 433.11402, 435.12967, 453.14023, 467.21339, 473.14532, 477.06746, 503.17701, 504.18751, 519.15079, 535.18209, 579.17192, 653.17232, 677.15119, 709.19853, 729.26113, 755.24040, 769.25605, 825.46419, 867.47475; P1: m/z 187.08658, 193.04953, 239.08150, 249.14852, 255.06518, 287.09140, 301.10705, 305.10196, 317.10196, 327.13393, 335.11252, 345.09687, 355.10235, 359.11252, 389.12309, 417.11800, 433.11292, 447.09218, 459.12857, 463.12348, 479.11840, 523.14461, 538.22828, 549.16026, 579.17083, 595.16574, 607.20213, 625.17631; P2: m/z 187.03897, 191.10665, 203.03388, 217.04953, 223.06009, 233.04444, 247.06009, 247.09648, 249.06585, 263.08150, 267.07640, 271.09648, 277.10705, 279.07641, 319.11761, 339.10744, 342.16998, 359.07614, 369.11800, 431.13365, 449.10783, 461.14422, 465.10275, 503.11840, 517.13405, 517.15518, 565.15518, 627.15557, 655.18687, 707.25215.
The high resolution extracted ion chromatograms (HREICs) of QQKDG in the positive (P) and negative ion modes (N). N1 m/z 179.03498, 191.01972, 299.05611, 315.05102, 353.08780, 359.07724, 375.12967, 403.12458, 415.10345, 435.15079, 461.10893, 461.16644, 507.11441, 515.11949, 519.18718, 535.16684, 579.20831, 609.14610, 623.19814, 719.16175; N2: m/z 177.01933, 191.05611, 253.07176, 269.04554, 285.04046, 301.03537, 329.06667, 361.11402, 367.10345, 373.11402, 387.16605, 389.10893, 431.09837, 449.14532, 461.07254, 473.07254, 477.10384, 477.14023, 491.11949, 515.14063, 547.14571, 593.15119, 609.18249, 637.10463; N3: m/z 197.04554, 239.09249, 283.15509, 311.04085, 315.07215, 335.13475, 341.08781, 357.13436, 445.07763, 445.11402, 447.09328, 447.15079, 463.08819, 481.09876, 505.15627, 507.17192, 521.13006, 563.14062, 607.20322, 621.18249, 623.16175, 639.19305, 717.14610; N4: m/z 169.01424, 207.06628, 223.09758, 267.06628, 271.06119, 283.06119, 295.04594, 299.11362, 327.05102, 329.08780, 331.06706, 337.09289, 339.07215, 344.04015, 349.08636, 355.10345, 373.09289, 373.12927, 387.14492, 407.09837, 433.11402, 435.12967, 453.14023, 467.21339, 473.14532, 477.06746, 503.17701, 504.18751, 519.15079, 535.18209, 579.17192, 653.17232, 677.15119, 709.19853, 729.26113, 755.24040, 769.25605, 825.46419, 867.47475; P1: m/z 187.08658, 193.04953, 239.08150, 249.14852, 255.06518, 287.09140, 301.10705, 305.10196, 317.10196, 327.13393, 335.11252, 345.09687, 355.10235, 359.11252, 389.12309, 417.11800, 433.11292, 447.09218, 459.12857, 463.12348, 479.11840, 523.14461, 538.22828, 549.16026, 579.17083, 595.16574, 607.20213, 625.17631; P2: m/z 187.03897, 191.10665, 203.03388, 217.04953, 223.06009, 233.04444, 247.06009, 247.09648, 249.06585, 263.08150, 267.07640, 271.09648, 277.10705, 279.07641, 319.11761, 339.10744, 342.16998, 359.07614, 369.11800, 431.13365, 449.10783, 461.14422, 465.10275, 503.11840, 517.13405, 517.15518, 565.15518, 627.15557, 655.18687, 707.25215.
3.2.1 Identification of flavonoids
Flavonoids are hydroxylated phenolic substances that exist as aglycones, glycosides or methylated derivatives as the secondary metabolites of plants in natural world (Harborne, 2013). The flavonoid could be subdivided into different subclasses according to the location of the B ring connection, the degree of unsaturation of the C ring and oxidation including isoflavones, flavones, flavonols, flavanones, dihydrochalcones, etc. (Corradini et al., 2011; Karak, 2019; Dias et al., 2021). In this study, a total of 89 flavonoids including 37 isoflavones, 29 flavones, 16 flavonols, 3 flavanones, 3 chromones, 1 dihydrochalcone, were detected and characterized in QQKDG based on FBMN and in-house library (Fig. 3). The identified isoflavones were derived primarily from Belamcanda chinensis and Puerariae lobatae, which are also a rich source of isoflavones and have an intense effect of anti-bacterial, anti-inflammatory, relieving pain (Wozniak et al., 2010; Choi et al., 2016). Reportedly, glycosyl group was easily substituted at the 7 or 4′ position of aglycones; consequently, the sugar moiety reduced to produce a distinct fragment ion at [Y0]± ion in O-glycosidic flavonoid compounds (Zhang et al., 2017). Furthermore, it is helpful in determining the presence of special functional groups in the structures in which the neutral loss of a molecule of H2O (18.011 Da), CO (27.995 Da), or CO2 (43.990 Da) and the cleavage of hexose occurred in the C-glycosidic flavonoid (Li et al., 2015). Peak 69 generated precursor ion [M−H]– at m/z 547.14571 (C26H27O13–), which produced the fragment ions at m/z 325.0700 [M−H−132.0417–90.0311]–, thus indicating the neutral elimination of the pentose moiety and characteristic cleavage of the 0/3 bond, 295.0609 [M−H−132.0417–120.0417]– by the loss of pentose moiety and cleavage of the 0/2 bond. Subsequently, an ion was obtained at m/z 267.0660 by the neutral loss of CO from m/z 295.060 in ESI– mode. Peak 69 also yielded a precursor ion [M + H]+ at m/z 549.16026 (C26H29O13+), which was fragmented into m/z 417.1164, corresponding to [M + H–132.0417]+, by the cleavage of pentose, m/z 297.0745 [M + H-132.0417–120.0417]+, m/z 399.1058 [M + H–132.0417–18.0100]+, and m/z 381.0953 [M + H–132.042–36.022]+ owing to the loss of one H2O and two H2O from m/z 417.1164, and m/z 351.0848 [M + H–C5H8O4-2H2O-OCH2]+ in ESI+ mode. Compared with the reference substance, peak 69 was identified as puerarin apioside. Peak 139 produced a precursor ion [M−H]– at m/z 461.10893 (C26H27O13–), which yielded fragment ions at m/z 299.0555 [M−H−C6H10O5]–, 284.0318 [M−H−C6H10O5−CH3]–, 283.0246 [M−H−C6H10O5−CH4]–, 256.0335 [M−H−C6H10O5−CH3−CO]–, and 240.0422 [M−H−C6H10O5−CH3−CO2]–. Peak 139 also exhibited an [M + H]+ ion at m/z 463.12348, along with MS2 fragments at m/z 301.0695 [M + H-C6H10O5]+ and 286.0458 [M + H-C6H10O5-CH3]+. Based on the mass of MS fragmentation and standards, this compound was identified as tectoridin. The [M−H]– ions of flavones, flavonols, flavanones, and dihydrochalcones underwent a neutral loss of CH3, CO, CO2, and H2O, along with a Retro-Diels-Alder (RDA) fragmentation reaction (Liu et al., 2005). Peak 99 generated a precursor ion [M−H]– at m/z 447.09328 (C21H19O11–), and was fragmented into m/z 285.0402 [M−H−162.0523]– by the loss of the C6H10O5 at the C-7 position, and into 151.0025 [C7H3O4]– and 133.0283 [C8H5O2]– by RDA cleavage. Thus, it was unambiguously characterized as luteolin-7-O-glucoside by comparing its retention time and fragment ions with the reference standard. The proposed fragmentation pathways for puerarin apioside, tectoridin, and luteolin-7-O-glucoside were shown in Fig. 4a, b, c, respectively.
The Feature-based molecular network of flavonoids of QQKDG extract in positive ion mode (a,b) and nagetive ion mode (c,d). a,c isoflavones; b,d flavones.
The Feature-based molecular network of flavonoids of QQKDG extract in positive ion mode (a,b) and nagetive ion mode (c,d). a,c isoflavones; b,d flavones.
The proposed fragmentation pathways of chemicals of QQKDG in positive and negative modes, a, puerarin apioside; b, tectoridin; c, luteolin-7-O-glucoside; d, forsythoside A; e, isochlorogenic acid A; f, isoimperatorin; g, rosmarinic acid; h, sweroside; i, magnoflorine.
The proposed fragmentation pathways of chemicals of QQKDG in positive and negative modes, a, puerarin apioside; b, tectoridin; c, luteolin-7-O-glucoside; d, forsythoside A; e, isochlorogenic acid A; f, isoimperatorin; g, rosmarinic acid; h, sweroside; i, magnoflorine.
The proposed fragmentation pathways of chemicals of QQKDG in positive and negative modes, a, puerarin apioside; b, tectoridin; c, luteolin-7-O-glucoside; d, forsythoside A; e, isochlorogenic acid A; f, isoimperatorin; g, rosmarinic acid; h, sweroside; i, magnoflorine.
3.2.2 Identification of phenylethanoid glycosides
Reportedly, the common chemical structures of phenylethanoid glycosides are composed of saccharides (including rhamnose, glucose, and phenethyl alcohol (C6–C2) moieties) and linked aromatic acids (including caffeic acid, coumaric acid, and ferulic acid) via glycosidic bonds (Zheng et al., 2014; Wang et al., 2019). Phenylethanoid glycosides, detected in QQKDG, are a type of signature ingredient and the main components that play therapeutic effects; they are rooted in Forsythia suspensa (Shao et al., 2017), Furthermore, phenylethanol glycosides can regulate various signaling pathways to play an antiinflammatory role, and are promising as a supplement for antiinflammatory drugs. In the ESI- mode, the primary and representative phenylethanoid glycosides lost were glucose (Glu C6H10O5, 162.0523 Da), rhamnose (Rha C6H10O4, 146.0574 Da), H2O (18.0101 Da), CO (27.9943 Da), and CO2 (43.9893 Da). In addition, a series of lower-molecular-weight aromatic acids showed regular fragmentation patterns. For instance, caffeic acid (C9H7O4–, m/z 179.0339) produced ions at m/z 135.0441 (C8H7O2–) and 161.02333 (C9H5O3–) by the further loss of CO2 and H2O; ferulic acid (C10H9O4–, m/z 193.0495) yielded an ion at m/z 175.0390(C10H7O3–) by the loss of H2O; and coumaric acid (C9H7O3–, m/z 163.0390) generated ions at m/z 145.0284 (C9H5O2–) and 119.0491 (C8H7O–) by the further loss of H2O and CO2, respectively. In this study, the FBMN contained a total of 91 nodes, of which numerous were phenylethanoid glycosides according to their fragment ions (Fig. 5); however, only a few nodes were analyzed and identified. Peaks 108, 117, 119, and 146, eluted at 10.05, 10.80, 10.95, and 12.54 min, respectively, showed the same precursor ion [M−H]– at m/z 623.19814 (C29H35O15–), and exhibited similar fragment ions at m/z 461.1665 [M−H−C6H10O5]–, 179.0340 [caffeic acid-H]–, and 161.0233 [caffeic acid-H2O]–. Peaks 117 and 119 were confirmed to be forsythoside. Additionally, peaks 108 and 146 were tentatively identified as forsythoside H and I, respectively, based on their retention behavior on a reversed-phase chromatographic column and similar fragmentation patterns (Sun et al., 2015). Peaks 106 and 132 (Rt 10.02 and 11.58 min, respectively) were assigned to calceolariosides A and B in the FBMN. The proposed fragmentation pathway of forsythoside A is shown in Fig. 4d.
The Feature-based molecular network of phenylethanoid glycosides of QQKDG extract in negative ion mode.
3.2.3 Identification of chlorogenic acid derivatives
Chlorogenic acids (CGAs) are a group of esters of hydroxycinnamic acids (HCAs), which include caffeic acid (CA), ferulic acid (FA), p-coumaric acid (p-CoA), and quinic acid (QA), which yield caffeoylquinic acid (CQA), feruloylquinic acid (FQA), and p-coumaroylquinic acid (pCoQA), respectively; additionally, the majority of CGAs were mono- and di-caffeoylquinic acids (CQAs) (Ramabulana et al., 2020). These compounds exist predominantly in several plants, which possess some notable pharmacological properties, such as antiinflammatory and antioxidant properties, and contribute significantly to the total dietary intake of phenols (Marques and Farah, 2009). The caffeic acid (m/z 179.0338 C9H7O4), ferulic acid (m/z 193.0495 C10H9O4), and p-coumaric acid (m/z 163.0390 C9H7O3) are the common types of hydroxycinnamic acids, which generate a series of characteristic fragment ions at m/z 135.0441 (C8H7O2–) and 161.0233 (C9H5O3–) by the further loss of CO2 and H2O from m/z 179.0338 (C9H7O4–); m/z 178.0261 (C9H6O4–), 149.0597 (C9H9O2–), and 134.0362 (C8H6O2–) by the further loss of CH3, CO2, CO2H2O from m/z 193.0495 (C10H9O4–), respectively; and 119.0491 (C8H7O–) by the loss of CO2 from m/z 163.0390. Meanwhile, the fragment ions at m/z 173.0444 (C7H9O5–), 127.0389 (C6H7O3–), and 111.0440 (C6H7O2–), by the loss of H2O, 2H2OCO, and 2H2OCO2 from m/z 11.0550 (C7H11O6–), respectively, were characteristic ions of quinic acid in the ESI– mode. In this study, the FBMN contained 28 nodes in the positive ion mode and 21 nodes in the negative ion mode in connection with CGAs; additionally, most of the nodes were hydroxycinnamylquinic acids and dihydroxycinnamylquinic acids (Fig. 6). Peaks 70, 138, 144, 145, and 162, eluted at 7.63, 12.05, 12.39, 12.52, and 13.04 min, respectively, exhibited the same precursor ion [M−H]– at m/z 515.11949 (C25H23O12–), which produced fragment ions at m/z 353.0874 [M−H−C9H6O3]– by the loss of a caffeoyl residue, 179.0340 [caffeic acid-H]–, 135.0439 [caffeic acid-H-CO2]–, 191.0552 [quinic acid-H]–, and 173.0445 [quinic acid-H-H2O]–. Peaks 70, 144, 145, and 162 were explicitly identified as 1,3-Dicaffeoylquinic acid, isochlorogenic acid B, isochlorogenic acid A, and isochlorogenic acid C, respectively, by comparison with corresponding reference standards. Peak 138 as its isomer had a similar precursor ion and fragmentation patterns and was tentatively identified as 1, 4-Dicaffeoylquinic acid according to its chromatographic retention behavior. The proposed fragmentation pathway of isochlorogenic acid A is shown in Fig. 4e.
The Feature-based molecular network of chlorogenic acid derivatives of QQKDG extract in positive (a) and negative (b) ion modes.
3.2.4 Identification of coumarins
Coumarins are common lactones formed by the intramolecular dehydration of cis-o-hydroxycinnamic acid and are characterized by a benzene ring attached to an alpha-pyrone ring. Coumarins are the most abundant components of Angelicae dahuricae (AD) in QQKDG and possess various pharmacological activities, including antioxidant, anticancer, and anticoagulant effects (Lu et al., 2020). The coumarins separated from AD are predominantly linear furocoumarins, which consist of a simple coumarin as the parent nucleus and a substituent group at the seven and six positions. Furthermore, the positions of the benzene rings of simple coumarins at the five, six, seven, and eight sites were replaced by hydroxyl, methoxyl, methylenedioxy, and isopentenyl groups (Zhu and Jiang, 2018). In ESI+ mode, the primary and representative losses of H2O (18.0100 Da), CH3 (15.0229 Da), CO (27.9943 Da), and CO2 (43.9893 Da) occurred in simple coumarins. In this study, FBMN contained 31 nodes, of which 22 were identified as coumarins and 5 were regarded as potential compounds according to their MS2 fragment ions (Fig. 7). Peak 195 showed a precursor ion [M + H]+ at m/z 317.10196 (C17H17O6+), which yielded product ions at m/z 233.0435 [M + H-C5H8O]+, 231.0279 [M + H-C5H10O]+, 218.0202 [M + H-C5H8O-CH3]+, 203.0332 [M + H-C5H10O-CO]+, and 175.0383 [M + H-C5H10O-2CO]+. Therefore, peak 195 was confirmed to be byakangelicol based on a retention time and fragmentation pathway similar to that of the reference standards. Peak 228 displayed the parent ion [M + H]+ at m/z 271.09648 (C16H14O4+), thus indicating the presence of characteristic fragment ions at m/z 203.0330 [M + H-C5H8]+, 175.0382 [M + H-C5H8-CO]+, and 147.0435 [M + H-C5H8-2CO]+. Thus, it was unambiguously identified as isoimperatorin by comparing the retention time and parent and fragment ions with those of the standard. Peaks 185 and 192 both showed quasi-molecular ion [M + H]+ at m/z 305.10196 (C16H17O6+) and produced a series of similar characteristic fragment ions at m/z 203.0330 [M + H-C5H10O2]+, 175.0383 [M + H-C5H10O2-CO]+, and 147.0434 [M + H-C5H10O2-2CO]+, which were tentatively characterized as prangenin hydrate, oxypeucedan hydrate, respectively, based on their different retention times in the HPLC chromatogram. The proposed fragmentation pathway of isoimperatorin is shown in Fig. 4f.
The Feature-based molecular network of coumarins of QQKDG extract in positive ion mode.
3.2.5 Identification of phenolic acids
Phenolic acids are a class of common compounds formed by the substitution of hydrogen atoms on benzene rings with carboxylic acid (—COOH) and hydroxyl groups (—OH), and are widely present in plants, plant foods, and human metabolites. Phenolic acids are excellent antioxidants that can alleviate physical damage and chronic diseases caused by free radicals (Chen et al., 2020). In this study, 36 phenolic acids were identified in QQKDG. However, no related FBMN cluster of phenolic acids was built in the GNPS, most likely because phenolic acids contain numerous varieties such as caffeic acid, gallic acid, protocatechuic acid, and ferulic acid, and no correlation exists between the secondary fragment ions of these compounds. Peak 164 exhibited a quasi-molecular ion [M−H]– at m/z 359.07724 (C18H15O8–), which produced the characteristic fragment ions at m/z 197.0446 [M−H−C9H6O3]– by the loss of the caffeoyl group, 179.0339 [M−H−C9H6O3−H2O]–, and 161.0233 [M−H−C9H6O3−2H2O]–, as shown in Fig. 8. Therefore, peak 164 was accurately identified as rosmarinic acid based on chromatographic information and fragmentation patterns of the reference substance. Peaks 6 and 49 were identified as gallic and caffeic acids, respectively. Peak 9 presented a [M−H]– ion at m/z 331.06706 (C13H15O10–), corresponding to gallic acid linked to a glucose, fragment ions at m/z 169.0112 [M−H−162.0522]– from gallic acid and m/z 125.0231 [M−H−162.0522–43.9893]– corresponding to the loss of CO2; hence, peak 9 was tentatively identified as gallic acid-4-O-glucoside. Peaks 21 and 35 displayed the precursor ion [M−H]– at m/z 341.08781(C15H17O9–), which yielded fragment ions at m/z 179.03440 [M−H−C6H10O5]– and 135.0439 [M−H−C6H10O5−CO2]– by the subsequent loss of CO2. Thus, they were tentatively characterized as caffeic acid-O-glucosides. Peak 83 showed a precursor ion [M−H]– at m/z 449.14532 (C22H25O10–) and then yielded product ions at m/z 315.1093 [M−H−C8H6O2]–, 193.0498 (C10H9O4–), 175.0390 [C10H9O4-H2O]–, and 165.0547 [C10H9O4-CO]–. According to the fragmentation patterns reported in the literature (Sun et al., 2015), peak 83 was tentatively identified as forsythenside F. The proposed fragmentation pathway of rosmarinic acid is shown in Fig. 4g.
The Feature-based molecular network of rosmarinic acid derivates of QQKDG extract in negative ion mode.
3.2.6 Identification of terpenoids and alkaloids
In this study, 14 terpenoids, including 10 monoterpenes, 2 sesquiterpenes, and 2 triterpenes, were identified according to the chromatographic retention behavior and fragmentation patterns of the in-house library. Peak 64 displayed a precursor ion [M + HCOOH-H]– at m/z 403.12458 (C17H23O11–), which generated the aglycone ion [M−H−C6H10O5]– at m/z 195.065 (C10H11O4–) by the neutral loss of a glucose; the subsequent loss of CO2 further produced the fragment ion [M−H−C6H10O5−CO2]– at m/z 151.0751 (C9H11O2–), and 125.0231 (C6H5O3–) originated from the RDA cleavage. Thus, it was identified as a sweroside by comparing the retention time and fragmentation pathways with those of the reference compound. Peaks 15, 29, and 31 all displayed the deprotonated molecular ion [M−H]– at m/z 375.12967 (C16H23O10–), and produced similar MS/MS spectrum patterns. They produced several fragment ions at m/z 213.0761 [M−H−C6H10O5]–, 169.0859 [M−H−C6H10O5−CO2]–, 151.0753 [M−H−C6H10O5−CO2−H2O]–, and 125.0595 (C7H9O2–), which were tentatively identified as 8-epiloganic acid (peak 15), loganic acid (peak 29), and adoxosidic acid (peak 21), according to the elution order on the chromatographic column and fragment pathways (He et al., 2020). Additionally, a total of 10 alkaloids that mainly originated from Isatis Tinctoria and Folium Isatidis in QQKDG were detected and characterized. Peak 56 was assigned to magnoflorine by comparing its retention time and MS2 spectrum with those of the reference compound. Peaks 222 and 225 showed a [M + H]+ ion at m/z 263.08150 (C16H11N2O2+), which produced fragment ions at m/z 235.0856 [M + H-CO]+ and 219.0907 [M + H-CO2]+, which were tentatively characterized as indigo and indirubin, respectively, according to the retention time and fragmentation patterns in the literature (Yan et al., 2017). The proposed fragmentation pathways for sweroside and magnoflorine are shown in Fig. 4h and i.
3.2.7 Identification of lignans and other compounds
Lignans are a class of natural products derived from the oxidative coupling of two C6–C3 units that can combine with sugar groups to form glycosides. Lignans have been used in traditional medicine for the treatment of diseases for a long time owing to their biological activities, including antioxidant, antitumor, anti-inflammatory (Teponno et al., 2016). In this study, 10 lignans were characterized using an in-house library and GNPS (Fig. 9). Peaks 148 and 171 both showed a quasi-molecular ion peak [M−H]– at m/z 519.18718 (C26H31O11–) and produced a similar fragment ion at m/z 357.1342 (C20H21O6–), corresponding to [M−H−C6H10O5]–, due to the loss of a glucose group. Hence, peaks 148 and 171 were tentatively characterized as pinoresinol-O-glucoside and matairesinoside, respectively, according to the elution order of the chromatographic column and fragment pathways (He et al., 2020). Additionally, the other seven compounds, namely Cgmp, N-octanoylsucrose, 3-(2′-Hydroxypheny)-4-(3H)-quinazolinone, ligustilide, Junipegenin C, pogostone, and kuzubutenolide A, were detected and putatively identified by matching the MS/MS product ions with in-house library. Peak 217 exhibited an [M + H]+ ion at m/z 191.10665, which produced product ions at m/z 145.1012 (C11H13–) owing to the loss of H2O and CO, and 117.0699 (C9H9–), derived from the ion at m/z 145.1012 through the elimination of 28.0305 (C2H4). It was identified as ligustilide by comparing the chromatographic behavior and MS/MS spectra with the reference standard.
The Feature-based molecular network of lignans of QQKDG extract in negative ion mode.
4 Conclusion
In this study, an integrated strategy based on UHPLC-Q-Exactive-MS coupled with FBMN analysis was developed for systematical characterization of structural types and identification of chemical ingredients in the TCM prescription QQKDG. It is beneficial to exploring unknown or potential compounds, revealing the visualization of the structural relationships among molecules, and improving the efficiency of compounds identification using multiple databases matching and fragmentation patterns, which could effectively avoid the problems of high cost and low efficiency of natural products discovery in TCMs. Thus, a total of 231 compounds were accurately or tentatively characterized, among which 224 compounds including ccflavonoids, phenolic acids, phenylethanoid glycosides, coumarins, chlorogenic acids, terpenoids, alkaloids, and lignans were identified for the first time. Moreover, numerous unassigned clusters and nodes were observed in the GNPS, which is conducive to the discovery of novel compounds. However, the results revealed that FBMN did not benefit the further characterization of isomers with high confidence, and distinguishing isomers is challenging at present. In summary, this systematic study on QQKDG provides a convenient and powerful analytical strategy for the rapid screening and detection of the chemical constituents of TCM formulas. The results of this study provide a theoretical basis for quality control and promote the development of modern QQKDG prescription.
Funding
This study was funded by the Science and Technology Innovation Program of Hunan Province (no. 2022RC1228), Hunan Province Social Science Innovation Research Base (Ethnic medicine and ethnic culture research base), Research Project of Sichuan Provincial Administration of Traditional Chinese Medicine (2021MS220) and Research Project of Hospital of Chengdu University of Traditional Chinese Medicine (20ZJ18).
Declaration of competing interest
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.
References
- Phytochemicals as potential antidotes for targeting NF-κB in rheumatoid arthritis. 3 Biotech.. 2017;7:1-11.
- [Google Scholar]
- Structure-antioxidant activity relationship of methoxy, phenolic hydroxyl, and carboxylic acid groups of phenolic acids. Sci. Rep.. 2020;10(1):2611.
- [Google Scholar]
- Systematic characterization of chemical constituents in Mahuang decoction by UHPLC tandem linear ion trap-Orbitrap mass spectrometry coupled with feature-based molecular networking. J. Sep. Sci.. 2021;44(14):2717-2727.
- [Google Scholar]
- Fermented Pueraria Lobata extract ameliorates dextran sulfate sodium-induced colitis by reducing pro-inflammatory cytokines and recovering intestinal barrier function. Lab Anim.. 2016;32:151-159.
- [Google Scholar]
- Flavonoids: chemical properties and analytical methodologies of identification and quantitation in foods and plants. Nat. Prod. Res.. 2011;25(5):469-495.
- [Google Scholar]
- Plant flavonoids: chemical characteristics and biological activity. Molecules.. 2021;26(17):5377.
- [Google Scholar]
- Qualitative analysis of chemical components in Lianhua Qingwen capsule by HPLC-Q Exactive-Orbitrap-MS coupled with GC-MS. J. Pharm. Sci.. 2021;11(6):709-716.
- [CrossRef] [Google Scholar]
- The integrated study on the chemical profiling and in vivo course to explore the bioactive constituents and potential targets of Chinese classical formula Qingxin Lianzi Yin Decoction by UHPLC-MS and network pharmacology approaches. J. Ethnopharmacol.. 2021;272:113917
- [Google Scholar]
- Harborne, J.B., 2013. The flavonoids: advances in research since 1980.
- Analysis of Chemical Material Basis of Compound Yuxingcao MixtureBased on HPLC-LTO Orbitrap MS/MS Technology. Asia-Pacific Tradit Med.. 2020;16(08):37-44.
- [Google Scholar]
- Design of new traditional Chinese medicine herbal formulae for treatment of type 2 diabetes mellitus based on network pharmacology. Chin. J. Nat. Med.. 2017;15(6):436-441.
- [Google Scholar]
- Biological activities of flavonoids: an overview. J. Pharm. Sci. Res.. 2019;10(4):1567-1574.
- [Google Scholar]
- Integrated molecular networking strategy enhance the accuracy and visualization of components identification: a case study of Ginkgo biloba leaf extract. J. Pharm. Biomed. Anal.. 2022;209:114523
- [Google Scholar]
- Preparative separation of isoflavones in plant extract of Pueraria lobata by high performance counter-current chromatography. J. Anal. Methods. 2015;7(4):1321-1327.
- [Google Scholar]
- Liquid chromatography/electrospray ionization mass spectrometry for the characterization of twenty-three flavonoids in the extract of Dalbergia odorifera. Rapid Commun. Mass Spectrom.. 2005;19(11):1557-1565.
- [Google Scholar]
- Traditional Chinese Medicine of Angelicae Pubescentis Radix: a Review of Phytochemistry. Pharmacol. Pharmacokinetics. Front. Pharmacol.. 2020;11
- [CrossRef] [Google Scholar]
- Chlorogenic acids and related compounds in medicinal plants and infusions. Food Chem.. 2009;113(4):1370-1376.
- [Google Scholar]
- The applications and features of liquid chromatography-mass spectrometry in the analysis of traditional Chinese medicine. Evid. Based Complementary Altern. Med. 2016
- [Google Scholar]
- Combining multidimensional chromatography-mass spectrometry and feature-based molecular networking methods for the systematic characterization of compounds in the supercritical fluid extract of Tripterygium wilfordii Hook F. Analyst. 2023;148(1):61-73.
- [Google Scholar]
- Profiling of chlorogenic acids from Bidens pilosa and differentiation of closely related positional isomers with the aid of UHPLC-QTOF-MS/MS-based in-source collision-induced dissociation. Metabolites. 2020;10(5):178.
- [Google Scholar]
- Forsythenethosides A and B: two new phenylethanoid glycosides with a 15-membered ring from Forsythia suspensa. Org. Biomol. Chem.. 2017;15(33):7034-7039.
- [Google Scholar]
- Comprehensive identification of 125 multifarious constituents in Shuang–huang–lian powder injection by HPLC-DAD-ESI-IT-TOF-MS. J. Pharm. Biomed. Anal.. 2015;115:86-106.
- [Google Scholar]
- Recent advances in research on lignans and neolignans. Nat. Prod. Rep.. 2016;33(9):1044-1092.
- [Google Scholar]
- An integrated approach to characterize intestinal metabolites of four phenylethanoid glycosides and intestinal microbe-mediated antioxidant activity evaluation in vitro using UHPLC-Q-Exactive High-Resolution Mass Spectrometry and a 1, 1-Diphenyl-2-picrylhydrazyl-based assay. Front. Pharmacol.. 2019;10:826.
- [Google Scholar]
- Identification of the effective constituents for anti-inflammatory activity of Ju-Zhi-Jiang-Tang, an ancient traditional Chinese medicine formula. J. Chromatogr. A.. 2014;1348:105-124.
- [Google Scholar]
- Rapid identification of chemical composition and metabolites of Pingxiao Capsule in vivo using molecular networking and untargeted data-dependent tandem mass spectrometry. Biomed. Chromatogr.. 2020;34(9):e4882.
- [Google Scholar]
- Identification of the absorbed components and metabolites of modified Huo Luo Xiao Ling Dan in rat plasma by UHPLC-Q-TOF/MS/MS. Biomed. Chromatogr.. 2018;32(6):e4195.
- [Google Scholar]
- Antimutagenic and anti-oxidant activities of isoflavonoids from Belamcanda chinensis (L.) DC. Mutat. Res. Genet. Toxicol. Environ. Mutagen.. 2010;696(2):148-153.
- [Google Scholar]
- Simultaneous determination of seven constituents in Oingqiaokangdu Granules by HPLC withswitching wavelengths. Chin Hosp. Pharm. J. Dec.. 2016;36(23):2065-2070.
- [CrossRef] [Google Scholar]
- The Clinical Study of Randomized Double Blind and Parallel Control, which is to Comparethe Efficacy between “Qingqiao Kangdu granules” Decocting-Free Granules and TraditionalDecoction in the Treatment of Common Cold with Exogenous Wind-Heat Syndrome. Chengdu Univ. Tradit. Chin. Med. 2015
- [Google Scholar]
- Observation on the curative effect of Qingqiao Kangdu granules intelligent free decoction in treating cold with wind-heat syndrome. Chin. J. Clin. Rational Drug Use. 2014;7(30):111-112.
- [CrossRef] [Google Scholar]
- Herb-drug interaction: an emerging issue of integrative medicine. Chinese J. Integrative Med.. 2010;16:195-196.
- [Google Scholar]
- Traditional Chinese medicine treatment of COVID-19. Complement. Ther. Clin. Pract.. 2020;39:101165
- [Google Scholar]
- Fingerprint Chromatograms and Chemical Componentsof Compound Indigowoad Root Granuleby APCI-MS and LC-APCI-MSn. J. Chin. Mass Spectrom. Soc.. 2017;38(03):320-327.
- [Google Scholar]
- A review of the ethnopharmacology, phytochemistry, pharmacology, application, quality control, processing, toxicology, and pharmacokinetics of the dried rhizome of Atractylodes macrocephala. Front. Pharmacol.. 2021;12:727154
- [Google Scholar]
- Effect of total flavonoids from astragalus complanatus on paraquat poisoning-induced pulmonary fibrosis in rats and its mechanisms. Chin J Ind. Hyg Occup. Dis.. 2015;33(11):838-840.
- [Google Scholar]
- An integrated approach for structural characterization of Gui Ling Ji by traveling wave ion mobility mass spectrometry and molecular network. RSC Adv.. 2021;11(26):15546-15556.
- [Google Scholar]
- An integrated strategy for the comprehensive profiling of the chemical constituents of Aspongopus chinensis using UPLC-QTOF-MS combined with molecular networking. Pharm Biol.. 2022;60(1):1349-1364.
- [Google Scholar]
- A novel isoflavone profiling method based on UPLC-PDA-ESI-MS. Food Chem.. 2017;219:40-47.
- [Google Scholar]
- Chemical and genetic discrimination of Cistanches Herba based on UPLC-QTOF/MS and DNA barcoding. PloS One.. 2014;9(5):e98061.
- [Google Scholar]
- Pharmacological and nutritional effects of natural coumarins and their structure–activity relationships. Mol. Nutr. Food Res.. 2018;62(14):1701073
- [Google Scholar]
Appendix A
Supplementary material
Supplementary data to this article can be found online at https://doi.org/10.1016/j.arabjc.2023.105463.
Appendix A
Supplementary material
The following are the Supplementary data to this article:Supplementary data 1
Supplementary data 1
Supplementary data 2
Supplementary data 2
Supplementary data 3
Supplementary data 3
Supplementary data 4
Supplementary data 4