An integrated strategy for comprehensive characterization of chemical components in Qingqiao Kangdu granules by UHPLC-Q-Exactive-MS coupled with feature-based molecular networking

Kailin Li; Ping Wu; Sanyu Li; Yongliang Huang; Ling Wang; Yifan Chen; Yufeng Zou; Fang Yan; Wei Cai

doi:10.1016/j.arabjc.2023.105463

An integrated strategy for comprehensive characterization of chemical components in Qingqiao Kangdu granules by UHPLC-Q-Exactive-MS coupled with feature-based molecular networking

Kailin Li^{^a,^b,1}, Ping Wu^{^c,1}, Sanyu Li^{^c,1}, Yongliang Huang^{^c}, Ling Wang^{^c}, Yifan Chen^{^b}, Yufeng Zou^{^b}, Fang Yan^{^a,⁎}, Wei Cai^{^b,⁎}

a Weifang Medical University, Weifang 261053, China

b Hunan Province Key Laboratory for Antibody-based Drug and Intelligent Delivery System, School of Pharmaceutical Sciences, Hunan University of Medicine, Huaihua 418000, China

c Hospital of Chengdu University of Traditional Chinese Medicine, Chengdu 610075, China

⁎Corresponding authors. yanfang303@163.com (Fang Yan), 20120941161@bucm.edu.cn (Wei Cai)

Received: 2023-7-18, Accepted: 2023-11-16,

Disclaimer:
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.

1 These authors contributed equally to the study.

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

Show Related Articles from PubMed

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

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

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

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

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

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

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

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).

Table 1 Identification of chemical components in QQKDG by UHPLC-Q-Exactive Orbitrap MS.

peak	t_R	Theoretical Mass m/z	Experimental Mass m/z	Error (ppm)	Formula	Identification	peak	t_R	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 595.16574	593.15088 595.16370	−0.53 −3.43	C₂₇H₃₀O₁₅	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 −3.46	C₂₁H₂₀O₁₂	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 625.17631	623.15979 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 −3.09	C₂₁H₂₀O₁₁	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 517.15518	515.14008 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 655.18687	653.17181 655.18469	−0.79–3.33	C₂₉H₃₄O₁₇	Iristectorigenin B-7-O-β-glucosyl (1 → 6) glucoside isomer
16	4.11	515.14063 517.15518	515.14001 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 517.15518	515.14014 517.15356	−0.95–3.13	C₂₂H₂₈O₁₄	caffeoylquinic acid-hexoside	138	12.05	515.11949 517.13405	515.11859	−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 −4.57	C₂₂H₂₂O₁₁	Tectoridin
24	4.82	461.16644	461.16586	−1.28	C₂₀H₃₀O₁₂	Forsythoside E	140	12.22	477.10384 479.11840	477.10379 479.11661	−0.12 −3.74	C₂₂H₂₂O₁₂	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 595.16574	593.15070 595.16364	−0.83 −3.53	C₂₇H₃₀O₁₅	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 479.11840	477.10342	−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 479.11840	477.10330	−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 −3.39	C₂₁H₁₈O₁₁	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 369.11800	367.10318 369.11664	−0.75 −3.70	C₁₇H₂₀O₉	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 595.16574	593.15100 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 415.10345	417.11606 415.10291	−4.67 −1.31	C₂₁H₂₀O₉	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 339.10744	337.09283 339.10574	−0.18 −5.02	C₁₆H₁₈O₈	p-Coumaroylquinic acid	182*	13.90	579.20831 [M−H + HCOOH]-	579.20770	−1.06	C₂₇H₃₄O₁₁	Forsythin
67	7.46	593.15119 595.16574	593.15100 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 549.16026	547.14539 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 655.18687	653.17212 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 655.18687	653.17151 655.18469	−1.24 −3.33	C₂₉H₃₄O₁₇	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 595.16574	593.15094 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 565.15518	563.14020 565.15283	−0.76 −4.16	C₂₆H₂₈O₁₄	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 369.11800	367.10297 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 565.15518	563.14038 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 −2.68	C₂₁H₂₀O₁₁	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 −3.647	C₂₁H₂₀O₁₁	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 479.11840	477.10364	−0.44	C₂₂H₂₂O₁₂	3′-Hydorxytectoridin	212	17.18	327.05102	327.05075	−0.84	C₁₇H₁₂O₇	Iriflogenin
97	9.51	563.14062 565.15518	563.14020	−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 −3.17	C₂₁H₂₀O₁₁	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 625.17631	623.16113 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 −2.88	C₂₁H₂₀O₁₂	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

* Compared with standard compounds.

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

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 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

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

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

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

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

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

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

Aravilli R.K., Vikram S.L., Kohila V., . Phytochemicals as potential antidotes for targeting NF-κB in rheumatoid arthritis. 3 Biotech.. 2017;7:1-11.
[Google Scholar]
Chen J., Yang J., Ma L., Li J., Shahzad N., Kim C.K., . Structure-antioxidant activity relationship of methoxy, phenolic hydroxyl, and carboxylic acid groups of phenolic acids. Sci. Rep.. 2020;10(1):2611.
[Google Scholar]
Chen L., Yao C., Li J., Wang J., Yao S., Shen S., . 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]
Choi S., Woo J.-K., Jang Y.-S., Kang J.-H., Jang J.-E., Yi T.-H., . 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]
Corradini E., Foglia P., Giansanti P., Gubbiotti R., Samperi R., Lagana A., . Flavonoids: chemical properties and analytical methodologies of identification and quantitation in foods and plants. Nat. Prod. Res.. 2011;25(5):469-495.
[Google Scholar]
Dias M.C., Pinto D.C., Silva A.M., . Plant flavonoids: chemical characteristics and biological activity. Molecules.. 2021;26(17):5377.
[Google Scholar]
Fu S., Cheng R., Deng Z., Liu T., . 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]
Gao W.-Y., Si N., Li M.-L., Gu X.-R., Zhang Y., Zhou Y.-Y., . 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.
He H.-H., Yiao J.-B., Chang W.-Q., Wang J.-F., Ying K., Luo T.-T., . 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]
Hu R.-F., Sun X.-B., . 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]
Karak P., . Biological activities of flavonoids: an overview. J. Pharm. Sci. Res.. 2019;10(4):1567-1574.
[Google Scholar]
Li Y., Cui Z., Li Y., Gao J., Tao R., Li J., . 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]
Li B., Li L., Zhao A., Han B., Fan Y., Liu C., . 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]
Liu R., Ye M., Guo H., Bi K., Guo D.a., . 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]
Lu Y., Wu H., Yu X., Zhang X., Luo H., Tang L., . Traditional Chinese Medicine of Angelicae Pubescentis Radix: a Review of Phytochemistry. Pharmacol. Pharmacokinetics. Front. Pharmacol.. 2020;11
[CrossRef] [Google Scholar]
Marques V., Farah A., . Chlorogenic acids and related compounds in medicinal plants and infusions. Food Chem.. 2009;113(4):1370-1376.
[Google Scholar]
Pang B., Zhu Y., Lu L., Gu F., Chen H., . The applications and features of liquid chromatography-mass spectrometry in the analysis of traditional Chinese medicine. Evid. Based Complementary Altern. Med. 2016
[Google Scholar]
Qu B., Liu Y., Shen A., Guo Z., Yu L., Liu D., . 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]
Ramabulana A.-T., Steenkamp P., Madala N., Dubery I.A., . 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]
Shao S.-Y., Feng Z.-M., Yang Y.-N., Jiang J.-S., Zhang P.-C., . 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]
Sun H., Liu M., Lin Z., Jiang H., Niu Y., Wang H., . 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]
Teponno R.B., Kusari S., Spiteller M., . Recent advances in research on lignans and neolignans. Nat. Prod. Rep.. 2016;33(9):1044-1092.
[Google Scholar]
Wang X., Chang X., Luo X., Su M., Xu R., Chen J., . 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]
Wang S., Chen P., Jiang W., Wu L., Chen L., Fan X., . 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]
Wang D., Fu Z., Xing Y., Tan Y., Han L., Yu H., . 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]
Wang N., Zhao X., Li Y., Cheng C., Huai J., Bi K., . 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]
Wozniak D., Janda B., Kapusta I., Oleszek W., Matkowski A., . 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]
Xia L.-L., Wang X.-H., Sheng R., . 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]
Xiong G.-F., . 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]
Xiong G.-F., Ynag T., Luo Y., Ma W.-Q., He Y., . 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]
Xu H., Chen K.-J., . Herb-drug interaction: an emerging issue of integrative medicine. Chinese J. Integrative Med.. 2010;16:195-196.
[Google Scholar]
Xu J., Zhang Y., . Traditional Chinese medicine treatment of COVID-19. Complement. Ther. Clin. Pract.. 2020;39:101165
[Google Scholar]
Yan J., Liu S., Li B.-P., Liu Z.-Q., Guo D.-F., . 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]
Yang L., Yu H., Hou A., Man W., Wang S., Zhang J., . 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]
Zhang Z., Dong Y., Huang Y., Peng L., . 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]
Zhang Y., Lei H., Tao J., Yuan W., Zhang W., Ye J., . 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]
Zhang F., Li B., Wen Y., Liu Y., Liu R., Liu J., . 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]
Zhang S., Zheng Z.-P., Zeng M.-M., He Z.-Y., Tao G.-J., Qin F., . A novel isoflavone profiling method based on UPLC-PDA-ESI-MS. Food Chem.. 2017;219:40-47.
[Google Scholar]
Zheng S., Jiang X., Wu L., Wang Z., Huang L., . Chemical and genetic discrimination of Cistanches Herba based on UPLC-QTOF/MS and DNA barcoding. PloS One.. 2014;9(5):e98061.
[Google Scholar]
Zhu J.J., Jiang J.G., . 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:

Fulltext Views
200

PDF downloads
111

View/Download PDF
Download Citations

BibTeX
RIS

Show Sections

[1] Aravilli R.K., Vikram S.L., Kohila V., . Phytochemicals as potential antidotes for targeting NF-κB in rheumatoid arthritis. 3 Biotech.. 2017;7:1-11.
[Google Scholar]

[2] Chen J., Yang J., Ma L., Li J., Shahzad N., Kim C.K., . Structure-antioxidant activity relationship of methoxy, phenolic hydroxyl, and carboxylic acid groups of phenolic acids. Sci. Rep.. 2020;10(1):2611.
[Google Scholar]

[3] Chen L., Yao C., Li J., Wang J., Yao S., Shen S., . 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]

[4] Choi S., Woo J.-K., Jang Y.-S., Kang J.-H., Jang J.-E., Yi T.-H., . 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]

[5] Corradini E., Foglia P., Giansanti P., Gubbiotti R., Samperi R., Lagana A., . Flavonoids: chemical properties and analytical methodologies of identification and quantitation in foods and plants. Nat. Prod. Res.. 2011;25(5):469-495.
[Google Scholar]

[6] Dias M.C., Pinto D.C., Silva A.M., . Plant flavonoids: chemical characteristics and biological activity. Molecules.. 2021;26(17):5377.
[Google Scholar]

[7] Fu S., Cheng R., Deng Z., Liu T., . 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]

[8] Gao W.-Y., Si N., Li M.-L., Gu X.-R., Zhang Y., Zhou Y.-Y., . 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]

[9] Harborne, J.B., 2013. The flavonoids: advances in research since 1980.

[10] He H.-H., Yiao J.-B., Chang W.-Q., Wang J.-F., Ying K., Luo T.-T., . 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]

[11] Hu R.-F., Sun X.-B., . 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]

[12] Karak P., . Biological activities of flavonoids: an overview. J. Pharm. Sci. Res.. 2019;10(4):1567-1574.
[Google Scholar]

[13] Li Y., Cui Z., Li Y., Gao J., Tao R., Li J., . 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]

[14] Li B., Li L., Zhao A., Han B., Fan Y., Liu C., . 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]

[15] Liu R., Ye M., Guo H., Bi K., Guo D.a., . 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]

[16] Lu Y., Wu H., Yu X., Zhang X., Luo H., Tang L., . Traditional Chinese Medicine of Angelicae Pubescentis Radix: a Review of Phytochemistry. Pharmacol. Pharmacokinetics. Front. Pharmacol.. 2020;11
[CrossRef] [Google Scholar]

[17] Marques V., Farah A., . Chlorogenic acids and related compounds in medicinal plants and infusions. Food Chem.. 2009;113(4):1370-1376.
[Google Scholar]

[18] Pang B., Zhu Y., Lu L., Gu F., Chen H., . The applications and features of liquid chromatography-mass spectrometry in the analysis of traditional Chinese medicine. Evid. Based Complementary Altern. Med. 2016
[Google Scholar]

[19] Qu B., Liu Y., Shen A., Guo Z., Yu L., Liu D., . 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]

[20] Ramabulana A.-T., Steenkamp P., Madala N., Dubery I.A., . 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]

[21] Shao S.-Y., Feng Z.-M., Yang Y.-N., Jiang J.-S., Zhang P.-C., . 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]

[22] Sun H., Liu M., Lin Z., Jiang H., Niu Y., Wang H., . 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]

[23] Teponno R.B., Kusari S., Spiteller M., . Recent advances in research on lignans and neolignans. Nat. Prod. Rep.. 2016;33(9):1044-1092.
[Google Scholar]

[24] Wang X., Chang X., Luo X., Su M., Xu R., Chen J., . 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]

[25] Wang S., Chen P., Jiang W., Wu L., Chen L., Fan X., . 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]

[26] Wang D., Fu Z., Xing Y., Tan Y., Han L., Yu H., . 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]

[27] Wang N., Zhao X., Li Y., Cheng C., Huai J., Bi K., . 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]

[28] Wozniak D., Janda B., Kapusta I., Oleszek W., Matkowski A., . 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]

[29] Xia L.-L., Wang X.-H., Sheng R., . 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]

[30] Xiong G.-F., . 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]

[31] Xiong G.-F., Ynag T., Luo Y., Ma W.-Q., He Y., . 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]

[32] Xu H., Chen K.-J., . Herb-drug interaction: an emerging issue of integrative medicine. Chinese J. Integrative Med.. 2010;16:195-196.
[Google Scholar]

[33] Xu J., Zhang Y., . Traditional Chinese medicine treatment of COVID-19. Complement. Ther. Clin. Pract.. 2020;39:101165
[Google Scholar]

[34] Yan J., Liu S., Li B.-P., Liu Z.-Q., Guo D.-F., . 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]

[35] Yang L., Yu H., Hou A., Man W., Wang S., Zhang J., . 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]

[36] Zhang Z., Dong Y., Huang Y., Peng L., . 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]

[37] Zhang Y., Lei H., Tao J., Yuan W., Zhang W., Ye J., . 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]

[38] Zhang F., Li B., Wen Y., Liu Y., Liu R., Liu J., . 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]

[39] Zhang S., Zheng Z.-P., Zeng M.-M., He Z.-Y., Tao G.-J., Qin F., . A novel isoflavone profiling method based on UPLC-PDA-ESI-MS. Food Chem.. 2017;219:40-47.
[Google Scholar]

[40] Zheng S., Jiang X., Wu L., Wang Z., Huang L., . Chemical and genetic discrimination of Cistanches Herba based on UPLC-QTOF/MS and DNA barcoding. PloS One.. 2014;9(5):e98061.
[Google Scholar]

[41] Zhu J.J., Jiang J.G., . Pharmacological and nutritional effects of natural coumarins and their structure–activity relationships. Mol. Nutr. Food Res.. 2018;62(14):1701073
[Google Scholar]

An integrated strategy for comprehensive characterization of chemical components in Qingqiao Kangdu granules by UHPLC-Q-Exactive-MS coupled with feature-based molecular networking

Abstract

Keywords

Qingqiao Kangdu granule

Characterization

UHPLC-Q-Exactive-MS

Feature-based molecular networking

1 Introduction

2 Materials and methods

2.1 Materials and reagents

2.2 Sample preparation

2.3 Liquid chromatographic conditions

2.4 MS spectrometry conditions

2.5 Feature-based molecular networking analysis

3 Results and discussion

3.1 Integrated strategy for data analysis

3.2 Characterization of chemical constituents in QQKDG by UHPLC-Q-Exactive-MS based on FBMN

3.2.1 Identification of flavonoids

3.2.2 Identification of phenylethanoid glycosides

3.2.3 Identification of chlorogenic acid derivatives

3.2.4 Identification of coumarins

3.2.5 Identification of phenolic acids

3.2.6 Identification of terpenoids and alkaloids

3.2.7 Identification of lignans and other compounds

4 Conclusion

Funding

Declaration of competing interest

References

Appendix A

Supplementary material

Appendix A

Supplementary material

Supplementary data 1

Supplementary data 2

Supplementary data 3

Supplementary data 4

Suggested read for related articles: