Rapid characterization of the chemical constituents of Sanhua decoction by UHPLC coupled with Fourier transform ion cyclotron resonance mass spectrometry.

Sanhua decoction, a famous Chinese herbal formula has been widely used for the treatment of stroke. In our study, a rapid, swift and straightforward analytical method with the help of UHPLC-FT-ICR-MS/MS was successfully developed for the first time to separate and identify the chemical constituents of Sanhua decoction. Chromatography was performed on a Universal XB C18 column (150 mm × 2.1 mm, 1.8 μm) using a mobile phase containing 0.1% formic acid-water (A) and acetonitrile (B). A total of 137 compounds in Sanhua decoction were identified or tentatively characterized. The findings revealed the fact that Sanhua decoction mainly contains flavonoids (in Aurantii fructus immaturus and Rheum palmatum L.), anthraquinones (in Rheum palmatum L.), coumarins (in Notopterygii Rhizoma Et Radix), phenylpropanoid glycosides, alkaloids and lignans (in Magnoliae Officmalis Cortex), which made up the key ingredients existing in Sanhua decoction. This study is hoped to be meaningful for the characterization of components in other traditional Chinese medicines, and lay the foundation for research on the pharmacology of Sanhua decoction. This journal is © The Royal Society of Chemistry.

Introduction

Traditional Chinese medicines (TCMs) play an important role in Chinese medicine due to their wide application and efficacy, particularly in the management of chronic diseases.[1] TCMs are composed of complex chemical constituents for multi-target treatment, and produce synergistic effects and reduce side effects. Although there has been a lot of research on TCMs, the active ingredients of most traditional Chinese medicine prescriptions remain unknown. Therefore, developing a fast and efficient analysis method to separate and identify complex constituents in TCMs is necessary for illustrating the pharmacological material basis of TCMs.

Sanhua decoction (SHD), a famous Chinese herbal formula, is recorded in the classic traditional Chinese medicine (TCM) book Suwen Bingji Qiyi Baomingji (Plain Questions: Discourse on Mechanism for Preserving Life). It consists of four crude herbs: Zhishi (Aurantii fructus immaturus), Dahuang (Rheum palmatum L.), Houpu (Magnoliae Officmalis Cortex) and Qianghuo (Notopterygii Rhizoma Et Radix). SHD is used widely for the treatment of stroke. It can improve ischemic cerebral edema and brain blood barrier (BBB) permeability.[2] It has been documented that Zhishi has neuroprotective, antioxidant, anti inflammatory and anti-apoptotic effects.[3] Dahuang can protect neurons from hypoxic-ischemic brain damage.[4] Houpu can protect neural damage from cerebral ischemia and reperfusion by suppressing cerebral inflammation and improving BBB function.[5] Qianghuo has a function of anti-thrombosis, increasing cerebral blood flow and improving cerebral blood circulation.[6] However, the chemical material basis of SHD has not been studied in detail until now.

As the complexity of traditional Chinese medicine, it is very important to establish a sensitive and reliable detection method. Chromatography-mass spectrometry (LC-MS) is widely used for its high sensitivity and specificity. Ultra high performance liquid chromatography (UHPLC) coupled with Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) as the most powerful instrument with high resolution, accuracy and sensitivity play a role in analyzing different kinds of complex samples.[7-10] It provides an unambiguous elemental composition and information about the isotopic abundance of ions to help determine the compositions and structures of chemical constituents. By usage of this technology, Li et al. characterized 179 constituents in Rhodiola crenulata as well as 37 prototypes and 142 metabolites in rats.[11] Guan et al. characterized 120 constituents in Sijunzi decoction[12] and Liu et al. characterized 174 constituents in Gegenqinlian decoction as well as 107 prototypes and 67 metabolites in rats.[13] The results show that this method is useful in detecting and indenting chemical constituents and metabolites by information about accurate molecular weight and MS2.

In this study, a rapid and simple approach by UHPLC-FT-ICR-MS was established for the systematical characterization of the constituents of SHD. This constituent characterization and structural elucidation provided significant information for a pharmacological study of SHD.

Material and methods

Chemicals and materials

The purchase of key ingredients including, Aurantii fructus immaturus (batch number: 170 901; source: Sichuan China), Rheum palmatum L. (batch number: 20 181 101; source: Gansu China), Magnoliae Officmalis Cortex (batch number: 181 001; source: Sichuan China) and Notopterygii Rhizoma Et Radix (batch number: 13 112 501; source: Sichuan China) was made from Guoda pharmacy (Shenyang, China), which followed identification by Professor Jingming Jia (Department of TCM, Shenyang Pharmaceutical University, Shenyang, China). The key source of the reference compounds (purity > 98%), including sennoside B, nodakenin, narirutin, sennoside A, neohesperidin, naringenin, hesperetin, nobiletin was Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China); naringin, hesperidin and honokiol was attained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). In addition, acetonitrile of HPLC grade, together with formic acid of LC-MS grade was attained from Fisher Scientific (Fair Lawn, NJ, USA), followed by attaining the purified water from Wahaha (Hangzhou, China).

Preparation of SHD for analysis

Following the documentary records of SHD, four constituting herbs that included Aurantii fructus immaturus (40 g), Rheum palmatum L. (40 g), Magnoliae Officmalis Cortex (40 g) and Notopterygii Rhizoma Et Radix (40 g) were soaked in water for 60 min, then decocted two times in 80 mL water for a period of 1 h each time in a glass flask and the ensuing solution was mixed and dried with the help of lyophilization. Prior to the analysis, dissolution the 0.5 g of dried powder was dissoluted by 5 mL methanol-aqueous (1 : 1) solution, soaking for 15 min with 80 °C water.

Instrument and analytical conditions

The sample was analyzed by the UHPLC-DAD-FT-ICR-MS system contains an Agilent 1260 UHPLC system combined with a Bruker Solarix7.0T FT-ICR-MS system (Germany) and Bruker Compass-Hystar workstation (Bruker, Germany). Chromatography was carried out on a Universal XB C18 column (150 mm × 2.1 mm, 1.8 μm; Kromat, USA) at the temperature of 35 °C. The mobile phase contained 0.1% formic acid–water (A) and acetonitrile (B). The gradient conditions were as follows: 8–12% (B) in 0 to 10 min, 12–13% (B) in 10 to 12 min, 13–18% (B) in 13 to 18 min, 18–24% (B) in 17 to 28 min, 24–30% (B) in 28 to 31 min, 30–36% (B) in 31 to 35 min, 36–55% (B) in 35 to 37 min, 55–90% (B) in 37 to 47 min, which was delivered at a flow rate of 0.2 mL min−1 and injection volume was 2 μL.

The mass spectra were performed with positive as well as negative electrospray ionization (ESI) modes, followed by setting the optimized conditions are as follows: capillary voltage, 4.5 kV; plate offset, 500 v; nebulizer gas pressure, 4.0 bar; dry gas flow rate, 8 L min−1; dry gas temperature, 200 °C; ion accumulation time, 0.15 s, and flight time, 0.6 ms. Full scan mass spectrometry data were recorded between m/z 100 and 1200 amu and the collision energy was ranged from 10 eV to 30 eV for MS/MS experiments.

Results and discussion

The Fig. 1 reveals the base peak ion chromatograms (BPC) of SHD in positive as well as negative ion modes, together with the respective compounds. The extract ion chromatograms (EIC) of each molecule weight were correspondingly attained for detecting the associated compound followed by presenting in support information Fig. S1.† Comparing with reference compounds, eleven peaks from SHD were accurately identified by the retention times and MS fragments in positive ion mode. The peak of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 in Fig. 1C were sennoside B, nodakenin, narirutin, naringin, sennoside A, hesperidin, neohesperidin, naringenin, hesperetin, honokiol, nobiletin, respectively. The others were predicted by Bruker workstation, whether or not the mass error values between the identified molecular weights with the measured molecular weights were below 3.0 ppm. If yes, it can infer the chemical structure of each component by the retention time and MS/MS data with published literature.[14-28] In the end, a total of 137 compounds were identified and characterized, whereas their layouts were presented in support information Fig. S2.†

Fig. 1

The base peak ion chromatograms (BPC) of SHT in both positive (A) and negative (B) ion modes and the corresponding compounds (C). (1) Sennoside B, (2) nodakenin, (3) narirutin, (4) naringin, (5) sennoside A, (6) hesperidin, (7) neohesperidin, (8) naringenin, (9) hesperetin, (10) honokiol, (11) nobiletin.

The MS/MS spectra of representative compounds are shown in Fig. 2, and displaying deduced fragmentation pathways are shown in Fig. 3. Except that, the formula, retention time, MS data, calculated m/z, error, ion mode and MS/MS data of other detected components were listed in Table 1.

Fig. 2

The MS/MS spectra of the typical compounds. (A) Hesperidin, (B) nomilin, (C) emodin-O-malonyl-glucose, (D) galloyl-glucoside, (E) epicatechin, (F) honokiol, (G) magnolol, (H) acteoside, (I) magnocurarine, (J) bergaptol.

Fig. 3

The possible fragmentation pathways of the typical compounds. (A) Hesperidin, (B) nomilin, (C) emodin-O-malonyl-glucose, (D) galloyl-glucoside, (E) epicatechin, (F) honokiol, (G) magnolol, (H) acteoside, (I) magnocurarine, (J) bergaptol.

UHPLC-FT-ICR-MS analysis of SHDa

No.	t R (min)	Identification	Formula	Monoisotopic mass (m/z)	Detected mass (m/z)	Ion mode	ppm	MS/MS (m/z)	Main source
1	2.07	Quinic acid	C7H12O6	193.07066	193.07024	[M + H]+	2.21	147.04051, 129.05446	C
2	2.64	Galloyl-glucoside	C13H16O10	331.06707	331.06759	[M − H]−	−1.57	313.05289, 271.04772, 211.02436, 169.01338, 125.00325	R
3	2.91	1-Galloyl-di-glucoside	C19H26O15	493.11989	493.12035	[M − H]−	−0.93	373.07705, 331.06771, 313.05667, 271.04642	R
4	3.21	Gallic acid	C7H6O5	169.01425	169.01447	[M − H]−	−1.30	151.00284, 125.02324	R
5	4.85	Catechin-3-O-glucoside	C21H24O11	451.12459	451.12518	[M − H]−	−1.31	289.07014, 245.08287	R
6	5.59	Rebouoside C	C20H30O12	461.16645	461.16674	[M − H]−	−0.64	299.11268, 137.05927	M
7	6.04	Chlorogenic acid	C16H18O9	353.08781	353.08824	[M − H]−	−1.23	191.09726, 179.76826, 173.10064	N
8	7.02	Catechin-5-O-glucoside	C21H24O11	451.12459	451.12504	[M − H]−	−1.01	289.07156, 245.08176, 203.07109	R
9	7.79	Magnocurarine	C19H24NO3+	314.17507	314.17473	[M + H]+	1.07	269.11618, 237.08974, 175.07514, 107.04887	M
10	10.40	Catechin	C15H14O6	289.07176	289.07211	[M − H]−	−1.22	245.08146, 205.04962, 203.06974	R
11	10.63	Cryptochlorogenic acid	C16H18O9	353.08781	353.08825	[M − H]−	−1.25	191.10029, 179.14054, 173.10065	N
12	10.64	Syringin	C17H24O9	395.13125	395.13094	[M + Na]+	0.80	372.08647, 343.94804, 208.09042, 137.03937	M
13	13.55	Magnoflorine	C20H24NO4+	342.16998	342.16967	[M + H]+	0.91	297.10956, 282.08696, 265.08507, 237.08972	M
14	16.29	(R)-Oblongine	C19H24NO3+	314.17507	314.17475	[M + H]+	1.02	269.11625, 237.09024, 175.07485, 107.04861	M
15	16.62	Epicatechin	C15H14O6	289.07176	289.07213	[M − H]−	−1.26	245.08235, 203.07153	R
16	18.49	Teupolioside	C35H46O20	785.25097	785.24931	[M − H]−	2.11	649.28262, 623.38565, 477.34832	M
17	19.22	Aloe-emodin-O-di-glucoside	C27H30O15	593.15119	593.15142	[M − H]−	−0.38	431.09851, 269.04559, 239.03448	R
18	19.31	Apigenin-6,8-di-C-glucoside	C27H30O15	595.16575	595.16502	[M + H]+	1.23	577.15636, 505.13492, 475.12405	C
19	20.32	Xanthoplanine	C21H25O4N+	356.18563	356.18528	[M + H]+	1.01	311.12628, 296.10825, 280.10864	M
20	20.74	Asimilobine	C17H17O2N	268.13321	268.13294	[M + H]+	0.99	251.10647, 236.08474, 219.08076, 191.08617	M
21	21.26	Echinacoside	C35H46O20	785.25097	785.24959	[M − H]−	1.76	649.28357, 623.38647, 477.34187	M
22	21.76	Chysoeriol-6,8-di-C-glucoside	C28H32O16	625.17631	625.17579	[M + H]+	0.84	607.16748, 535.14662, 505.13284	C
23	21.95	1-Cinnamoyl-6-O-(glucosyl)-galloyl-glucoside	C28H32O16	623.16176	623.16247	[M − H]−	−1.15	461.10814, 331.06702, 313.05827, 211.02305	R
24	22.31	Scopoletin	C10H8O4	193.04954	193.04935	[M + H]+	0.95	178.02482, 165.05373, 122.03784	C
25	22.45	2-O-Cinnamyl-β-d-glucoside	C15H18O7	309.09798	309.09841	[M − H]−	−1.39	237.07692, 201.01967, 146.96590	R
26	22.55	Emodin-O-glucuronic acid methyl ester	C22H20O11	459.09329	459.09358	[M − H]−	−0.65	269.04543, 241.05106, 225.05490, 213.05525	R
27	23.60	Decuroside V	C20H20O10	447.12617	447.12576	[M + Na]+	0.91	263.08937, 245.08472	N
28	23.73	Xanthotoxol	C11H6O4	203.03389	203.03379	[M + H]+	0.45	173.07564, 145.04536, 117.05273	C
29	23.73	Bergaptol	C11H6O4	201.01933	201.01957	[M − H]−	−1.17	185.02456, 157.17256, 129.10451, 101.96320	N
30	23.76	1-O-Cinnamyl-β-d-glucoside	C15H18O7	309.09798	309.09843	[M − H]−	−1.47	248.95978, 161.04574, 146.96419	R
31	23.77	Bergaptol-O-β-d-glucopyranoside	C17H16O9	365.08671	365.08641	[M + H]+	0.81	201.03287, 159.17264, 131.10514	N
32	24.94	Natsudaidain-3-O-glucoside	C27H32O14	581.18648	581.18599	[M + H]+	0.85	419.13174, 273.07487, 153.01846	C
33	24.97	Narirutin-4′-O-glucoside	C33H42O19	743.23931	743.23875	[M + H]+	0.75	581.18682, 435.12794, 273.07526	C
34	25.16	Eriocitrin	C27H32O15	597.18140	597.18096	[M + H]+	0.73	435.12791, 289.06947, 153.01849	C
35	25.35	Resveratrol-4′-O-glucoside	C20H22O8	389.12419	389.12479	[M − H]−	−0.83	245.08196, 227.07138, 199.07567, 135.04384	R
36	25.47	Isolindleyin	C23H26O11	477.14024	477.14067	[M − H]−	−0.90	313.05691, 211.02328, 169.01307, 125.02387	R
37	25.79	Lindleyin	C23H26O11	477.14024	477.14071	[M − H]−	−1.12	313.05659, 211.02454, 169.01340, 125.02321	R
38	26.19	Limonin	C26H30O8	471.20134	471.20060	[M + H]+	0.83	425.19343, 339.19426, 161.05897	C
39	26.19	Rhein-8-O-β-d-glucoside	C21H18O11	445.07763	445.07816	[M − H]−	−1.21	285.07559, 267.03326, 257.07879, 241.07041	R
40	26.28	Nicotiflorin	C27H30O15	595.16575	595.16515	[M + H]+	1.00	419.13403, 390.09214, 287.07785	C
41	26.28	Naringenin-7-O-glucoside	C21H22O10	435.12857	435.12833	[M + H]+	0.56	272.07462, 227.18968, 177.22102, 165.13608	C
42	26.29	Eriodictyol	C15H12O6	289.07066	289.07051	[M + H]+	0.53	153.01692, 135.06538, 107.04796	C
43	26.42	Limonin-17-β-d-glucoside	C32H42O14	649.25018	649.25108	[M − H]−	−1.39	487.19725, 425.18954, 339.18671	C
44	26.49	Neoeriocitrin	C27H32O15	597.18140	597.18134	[M + H]+	0.52	435.12786, 289.06814, 153.01794	C
45	26.55	Isosakuranetin	C16H14O5	287.09140	287.09138	[M + H]+	0.38	153.01718, 133.065095	C
46	27.07	Alhagidin	C34H44O20	771.23532	771.23656	[M − H]−	−1.61	635.18786, 609.19823, 161.02394	M
47	27.17	Cinnamoyl-6-O-galloyl-diglucoside	C28H32O16	623.16176	623.16185	[M − H]−	−1.50	475.10891, 331.06765, 313.05664	R
48	27.25	Nodakenin	C20H24O9	407.13476	407.13536	[M − H]−	−1.48	247.04625, 229.14536, 175.41575	N
49	27.29	Marmesin	C14H14O4	247.09649	247.09655	[M + H]+	−0.28	153.01714	C
50	27.38	Veronicastroside	C27H30O15	595.16575	595.16502	[M + H]+	0.84	577.15147, 457.10763, 295.05903	C
51	27.40	Sennoside B	C42H38O20	861.18837	861.18831	[M − H]−	−1.71	699.13514, 655.14385, 389.08791	R
52	27.44	Emodin-O-di-glucoside	C27H30O15	593.15119	593.15112	[M − H]−	−1.24	269.04549, 240.04318, 225.05493	R
53	28.01	Cassiachromone	C13H12O4	231.06628	231.06653	[M − H]−	−1.09	269.04549, 240.04321	R
54	28.96	Narirutin	C27H32O14	581.18648	581.18594	[M + H]+	0.94	435.12779, 419.13317, 273.07528	C
55	29.56	Isorhoifolin	C27H30O14	579.17083	579.17056	[M + H]+	0.47	449.14346, 303.08368, 153.01728	C
56	29.64	Chrysophanol-1,8-O-di-glucoside	C27H30O14	577.15628	577.15658	[M − H]−	−0.52	415.10387, 253.04962, 225.05395	R
57	30.04	Acteoside	C29H36O15	623.19814	623.19902	[M − H]−	−0.36	461.16314, 161.02265	M
58	30.15	1-Dihydro-p-coumaroyl-6-O-galloyl-glucoside	C22H24O12	479.11950	479.12006	[M − H]−	−1.16	461.10874, 313.05591, 169.01294, 125.02296	R
59	30.21	Prunin	C21H22O10	435.12857	435.12887	[M + H]+	−0.68	273.07329, 153.01674	C
60	30.22	Natsudaidai	C21H22O9	419.13366	419.13349	[M + H]+	0.39	390.09212, 389.09134, 371.08154	C
61	30.36	Naringin	C27H32O14	581.18648	581.18605	[M + H]+	0.75	435.12862, 419.13274, 273.07564	C
62	30.80	Sennoside A	C42H38O20	885.18486	885.18538	[M + Na]+	−0.59	701.13594, 657.14538, 391.08807, 270.03807	R
63	30.94	Isoacteoside	C29H36O15	623.19814	623.19925	[M − H]−	−1.77	461.16741, 161.02568	M
64	30.98	Laccaic acid-d-O-glucoside	C22H20O12	475.08740	475.08820	[M − H]−	−0.91	431.09851, 269.03452, 253.05063, 240.04252	R
65	31.45	Rhoifolin	C27H30O14	579.17083	579.17057	[M + H]+	0.57	433.10764, 271.05967, 153.02389	C
66	31.45	Neodiosmin	C28H32O15	609.18140	609.18107	[M + H]+	0.77	301.07091, 286.04659	C
67	31.48	Isosakuranin	C22H24O10	449.14422	449.14408	[M + H]+	0.96	301.06877, 287.09012, 286.04485	C
68	31.59	Hesperidin	C28H34O15	611.19705	611.19650	[M + H]+	0.10	449.14312, 303.08581, 288.09732, 153.02175,	C
69	32.03	Nodakenetin	C14H14O4	245.08193	245.08221	[M − H]−	−1.14	229.10047, 211.64807, 175.42487	N
70	32.36	Physcion-1-O-glucuronic acid	C22H20O11	459.09329	459.09368	[M − H]−	−0.86	283.06124, 268.03815, 212.04737, 184.05123	R
71	32.78	Diosmin	C28H32O15	607.16684	607.16710	[M − H]−	−0.42	299.10025, 284.29568	C
				609.18140	609.18081	[M + H]+	0.96	301.06863 286.04525
72	32.89	Neohesperidin	C28H34O15	611.19705	611.19637	[M + H]+	0.53	449.14228, 303.08417, 153.01734	C
73	32.89	Hesperetin-7-O-glucoside	C22H24O11	465.13914	465.13958	[M + H]+	−0.92	303.08652, 288.03284, 153.01532	C
74	32.90	Sakuranetin	C22H24O10	449.14422	449.14477	[M + H]+	−1.22	301.06945, 287.08765, 286.045473	C
75	32.97	Magnoloside E	C28H34O15	609.18249	609.18261	[M − H]−	0.19	447.15213, 315.11019, 161.02613	M
76	33.05	Homoeriodictyol	C16H14O6	303.08631	303.08630	[M + H]+	0.06	288.04856, 153.01712, 117.03375	C
77	33.18	Limocitrin-3-O-(3-hydroxy-3-methylglutarate)-glucoside	C29H32O17	653.17123	653.17394	[M + H]+	−0.73	539.13891, 347.07494	C
78	33.40	1-Dihydrocinnamoyl-6-O-galloyl-glucoside	C22H24O11	463.12459	463.12503	[M − H]−	−0.97	331.06693, 313.05656, 211.02564, 169.01332	R
79	34.30	Magnolignan B	C18H20O5	315.12380	315.12417	[M − H]−	−0.95	267.10167, 249.09178, 221.09658, 133.06619	M
80	34.51	Meranzin	C15H16O4	261.11214	261.11203	[M + H]+	0.42	243.10142, 159.04375, 103.05456	C
81	34.59	Hesperidone hydrate	C15H18O5	296.14925	296.14925	[M + HH4]+	0.00	403.12656, 418.12351, 165.05352	C
82	35.89	Diosmetin-7-O-(6′′-O-acetyl) neohesperidoside	C30H34O17	667.18688	667.18653	[M + H]+	−0.26	301.07182, 286.05412	C
83	36.14	Natsudaidain-3-O-(3-hydroxy-3-methylglutarate)-glucoside	C33H40O18	725.22874	725.22824	[M + H]+	0.47	581.18621, 419.13224	C
84	36.29	Magnolignan A	C18H20O4	299.12888	299.12924	[M − H]−	−1.18	239.10865, 221.10746	M
85	37.01	Physcion-8-O-glucuronic acid	C22H20O11	459.09329	459.09369	[M − H]−	−0.89	283.06138, 268.03804, 239.03501, 224.04837	R
86	38.08	Cinnamoyl-6-O-galloyl-glucoside	C22H22O11	461.10894	461.10948	[M − H]−	−1.18	401.08794, 313.05874, 285.03987, 271.04612	R
87	39.00	Didymin	C28H34O14	595.20213	595.20196	[M + H]+	0.70	433.14783, 287.09001, 153.01794	C
88	39.08	Beta-hydroxyacteoside	C29H36O16	639.19306	639.19315	[M − H]−	−0.14	503.14287, 477.16173, 161.02408, 133.03174	M
89	39.39	6-Methoxy-8-O-β-d-glucoside	C20H24O9	407.13476	407.13521	[M − H]−	−1.12	253.04989, 201.09011, 146.96598	R
90	39.47	Magnoloside A	C29H36O15	623.19705	623.19645	[M + H]+	0.95	461.16607, 161.02574	M
91	39.51	Chrysophanol isomer	C15H10O4	253.05063	253.05093	[M − H]−	−1.23	225.05517, 210.03072, 197.06054	R
92	39.52	Chrysophanol-1-O-glucoside	C21H20O9	415.10346	415.10402	[M − H]−	−1.35	253.04884, 225.05627, 209.05942	R
93	39.70	Emodin-O-glucoside	C21H20O10	431.09837	431.09867	[M − H]−	−0.7	269.04562, 241.04997, 225.05524	R
94	39.74	3,5,6,7,8,3′,4′-Heptamethoxy flavone	C22H24O9	433.14931	433.14910	[M + H]+	0.49	418.12356, 403.12656, 165.05354	C
95	39.79	1-P-Coumaroyl-6-O-galloyl-di-glucoside	C29H36O16	639.19306	639.19367	[M − H]−	−0.96	477.13972, 313.05688, 211.02412, 141.03624	R
96	39.87	Sakuranin	C16H14O5	287.09142	287.09127	[M + H]+	0.46	153.01721, 133.06492	C
97	39.88	Poncirin	C28H34O9	595.20213	595.20168	[M + H]+	0.73	433.14812, 287.09110, 153.16712	C
98	39.93	Magnoloside M	C29H36O15	623.19705	623.19643	[M + H]+	0.99	461.16597, 161.02576	M
99	39.98	6′-O-trans-Feruloylnodakenin	C30H32O12	583.18213	583.18148	[M − H]−	1.07	247.10278, 229.44072, 175.28676	N
				607.17865	607.17861	[M + Na]+	−0.02	249.11947, 177.29759
100	40.24	Chrysophanol	C15H10O4	253.05063	253.05094	[M − H]−	−1.21	225.05537, 210.03104, 202.88598	R
101	40.93	Marmin	C19H24O5	333.16965	333.16942	[M + H]+	0.70	163.03837, 107.08582	C
102	41.15	2-(2′-hydroxypropyl)-5-methyl-7-hydroxychromone	C13H14O4	233.08193	233.08217	[M − H]−	−1.01	189.05568, 146.96597, 129.97617	R
103	41.29	Magnolignan C	C18H20O4	299.12888	299.12928	[M − H]−	−1.33	258.09247, 239.10852, 197.06244	M
104	41.55	Naringenin	C15H12O5	273.07575	273.07562	[M + H]+	0.48	153.17823, 133.06712	C
105	42.12	Procyanidin B	C30H26O12	577.13515	577.13535	[M − H]−	−0.49	425.08747, 407.07637, 289.07107	R
106	42.26	Physcion-8-O-glucuronic acid methyl ester	C23H22O11	473.10894	473.10946	[M − H]−	−1.11	283.06094, 268.03811, 240.04325	R
107	42.27	Emodin-O-malonyl-glucose	C24H22O13	517.09876	517.09935	[M − H]−	−1.15	473.10884, 431.09157, 269.04531, 241.05131	R
108	42.51	Chrysophanol-8-O-glucoside	C21H20O9	415.10346	415.10400	[M − H]−	−1.30	253.04984, 225.05618, 209.05959	R
109	42.51	Emodin	C15H10O5	269.04555	269.04588	[M − H]−	−1.22	241.05016, 225.05524, 197.06027	R
110	42.53	Aloe-emodin-O-glucoside	C21H20O10	431.09885	431.09886	[M − H]−	−1.14	269.04462, 239.03647, 211.02524	R
111	42.61	Physcion-1-O-glucoside	C22H22O10	445.11402	445.11460	[M − H]−	−1.30	427.03257, 283.06205, 269.04607	R
112	42.61	Physcion	C16H12O5	283.06120	283.06162	[M − H]−	−1.49	283.07696, 268.04557, 240.04266	R
113	42.62	Hesperetin	C16H14O6	303.08631	303.08608	[M + H]+	0.77	153.01741, 110.04783	C
114	42.85	Magnatriol	C15H14O3	241.08702	241.08730	[M − H]−	−1.17	223.07597, 197.09716, 133.06719	M
115	42.97	Magnolignan E	C18H18O4	297.11323	297.11365	[M − H]−	−1.39	225.09132, 184.05786, 183.04759	M
116	43.29	5-Hydroxy-6,7,3′,4′,5′-pentamethoxyflavone	C20H20O8	389.12309	389.12283	[M + H]+	0.67	374.09631, 359.08221, 356.08174, 328.07931	C
117	43.54	5-o-Demethylnobiletin	C20H20O7	373.12818	373.12781	[M + H]+	0.73	358.10001, 343.07224, 325.06554, 297.06973	C
118	43.78	Demethylnobiletin	C20H20O8	389.12309	389.12282	[M + H]+	0.79	359.08112, 360.07831, 341.06984	C
119	44.32	Obovatol	C18H18O3	281.11832	281.11866	[M − H]−	−1.20	267.10174, 249.08513	M
120	44.54	Sinensetin	C20H20O7	373.12818	373.12790	[M + H]+	0.74	175.01549, 147.03194, 119.01593	C
121	44.59	Honokiol	C18H18O2	265.12340	265.12413	[M − H]−	−0.42	224.08126, 223.07514, 197.05923	M
122	44.59	O-Prenyl-umbelliferone	C14H14O3	229.08072	229.08726	[M − H]−	−1.04	160.01707, 142.17814	N
123	44.74	Aloe-emodin	C15H10O5	269.04555	269.04598	[M − H]−	−1.60	253.05121, 240.04247, 211.03968	R
124	44.94	Magnaldehyde D	C16H14O3	253.08702	253.08735	[M − H]−	−1.32	235.07963, 207.08634, 194.04012	M
125	45.13	Phelloptein	C17H16O5	323.08899	323.08875	[M + Na]+	0.76	269.07627, 231.01796	N
126	45.25	Notopterol	C21H22O5	353.13945	353.13988	[M − H]−	−1.21	203.17578, 159.09786, 147.63478	N
				377.13594	377.13559	[M + Na]+	0.95	205.20652, 161.11072
127	45.38	Rhein	C15H8O6	283.02481	283.02504	[M − H]−	−0.82	257.04507, 239.03461, 211.03958	R
128	45.38	Danthron	C14H8O4	239.03498	239.03530	[M − H]−	−1.33	186.73096, 141.51394, 130.71196	R
129	45.58	Nobiletin	C21H22O8	403.13874	403.13819	[M + H]+	1.41	373.09359, 355.08584, 327.08476	C
130	45.69	Nomilin	C28H34O9	515.22756	515.22753	[M + H]+	0.05	469.21806, 455.20675, 411.21462, 161.05854	C
131	45.88	Notoptol	C21H22O5	353.13945	353.13987	[M − H]−	−1.19	225.17472, 207.13978, 189.14075	N
				377.13594	377.13563	[M + Na]+	0.97	227.19076, 209.14289
132	46.17	Magnaldehyde E	C16H14O3	253.08702	253.08735	[M + H]+	−1.32	225.09424, 184.05624, 183.05074	M
133	46.17	Magnolol	C18H18O2	265.12340	265.12394	[M − H]−	−0.39	247.10096, 245.09517, 223.07491	M
134	46.19	p-Hydroxyphenethyl anisate	C16H16O4	295.09408	295.09386	[M + Na]+	0.73	153.07368, 138.08654	N
135	46.54	Obovaaldehyde	C16H14O4	269.08193	269.08215	[M − H]−	−0.80	152.01478, 124.02018	M
136	46.70	Isosinensetin	C20H20O7	373.12818	373.12783	[M + H]+	0.94	358.10398, 343.08158, 315.08276	C
137	46.70	Tangeretin	C20H20O7	395.11012	395.10998	[M + Na]+	0.35	278.54947, 276.55982, 243.12708	C

Ps: C represented Aurantii fructus immaturus; R represented Rheum palmatum L.; M represented Magnoliae Officmalis Cortex; N represented Notopterygii Rhizoma Et Radix.

Characterization of flavonoids in SHD

Flavonoids have a typical cyclohexene structure in the C ring, which is unstable and easy to break it for Diels–Alder reaction (RDA), and get the characteristic fragment ions 1,3A and 1,3B. Flavonoid glycosides have methoxy group, firstly losing the glycosyl groups, followed by losing the methyl groups, at last cleavage of the C ring. The mass spectral fragmentation of flavonoids by hesperidin as an example was described. In positive ion mode, adduct ion at m/z 611.19650 ([M + H]+) indicating an elemental composition of C28H34O15. The MS/MS spectrum (Fig. 3A) is presented that the main fragment ions are m/z 449.14312, 303.08581, 288.09732, 153.02175 as well as 136.04821. Among them, m/z 449.14312 and 303.08581 are inferred by cleaving of the glycosidic linkage and losing C6H10O5 (162 Da) or C12H20O9 (308 Da) from the ion of [M − H]+, m/z 288.09732 is the result of losing CH3 (15 Da) from m/z 303.08581, and m/z 153.02175 and m/z 136.04821 was due to the Diels–Alder reaction from m/z 288.09732. The characterization of other flavonoids in SHD was performed based on the fragmentation patterns and related literatures.[14-18]

Characterization of triterpenoids in SHD

Triterpenoids are prone to lose H2O (18 Da), CO (28 Da) and CO2 (44 Da), due to the presence of lactone rings. Nomilin was selected to illustrate the fragmentation pathways. In positive ion mode, adduct ion at m/z 515.22753 ([M + H]+) indicating an elemental composition of C28H34O9. The MS/MS spectrum (Fig. 3B) is presented that the main fragment ions are m/z 469.21806, 455.20675, 411.21462, 161.05854. Among them, m/z 469.21806 is inferred by losing CO and H2O (46 Da) from [M + H]+, m/z 455.20675 is presumed by losing C2H4O2 (60 Da) from [M + H]+, m/z 411.21462 is the result of losing CO2 (44 Da) from m/z 455.20675, and m/z 161.05854 is estimated by losing C18H26O7 (354 Da) from [M + H]+. The characterization of other triterpenoids in SHD was performed based on the fragmentation patterns and related literatures.[15,19]

Characterization of anthraquinones in SHD

The common characteristic of MS2 in anthraquinones is the continuous loss of CO (28 Da). Anthraquinone glycosides can lose glycosyl and CO, continually. Aglycones can be judged by [aglycone − H]−. Anthraquinone glycosides can lose methyl-glucuronide (190 Da), cinnamoyl (148 Da), malonyl (86 Da), acetyl-glucose (204 Da), and glucose (162 Da). Emodin-O-malonylglucose was selected to illustrate the fragmentation pathways. In negative ion mode, adduct ion at m/z 517.09935 ([M − H]−) indicating an elemental composition of C24H22O13. The mass spectrum (Fig. 3C) shows that the main fragment ions are m/z 473.10884, 431.09157, 269.04531 and 241.05131. Among them, m/z 473.10884 is presumed by losing of CO2 (44 Da) from [M − H]−, m/z 431.09157 is inferred by losing of the C3H2O3 (86 Da) from [M − H]−, and m/z 269.04531 is speculated by cleaving of the glycosidic linkage and losing C6H10O5 (162 Da) from m/z 431.09157, m/z 241.05131 is the result of losing CO (28 Da) at m/z 269.04531. The characterization of anthraquinone compounds in SHD is inferred from this cleavage rule and related literatures.[20,21]

Characterization of glucosides in SHD

Glucosides are formed by the esterification of acid with glucose. According to the type of acid, it can be divided into cinnamoyl-glucose, dihydrocinnamoyl-glucose, p-cinnamoyl-glucose, dihydro-p-cinnamoyl glucose, galloyl-glucose, and resveratrol-galloyl-glucose in rhubarb. They are vulnerable to loss of neutral fragments 148 Da, 150 Da, 164 Da, 166 Da, 170 Da, and 228 Da, respectively. Galloyl-glucose was selected to illustrate the fragmentation pathways. In negative ion mode, adduct ion at m/z 331.06759 ([M − H]−) indicating an elemental composition of C13H16O10. The mass spectrum (Fig. 3D) shows that the main fragment ions are m/z 313.05289, 271.04772, 211.02436, 169.01338, and 125.00325. Among them, m/z 313.05289 is presumed by losing H2O (18 Da) from [M − H]−, m/z 271.04772 is inferred by losing C2H4O2 (60 Da) from [M − H]−, m/z 211.02436 is the result of losing C4H8O4 (120 Da) from [M − H]−, m/z 169.01338 is speculated by cleaving of the glycosidic linkage and losing glucose C6H10O5 (162 Da) from [M − H]−, and m/z 125.00325 is deduced by losing CO2 (44 Da) from m/z 169.01338. The characterization of glucosides in SHD is concluded by referring to the cleavage rule and related literature.[21]

Characterization of epicatechin in SHD

Epicatechin was selected to illustrate the fragmentation pathways for catechins. The adduct ion at 289.07213 ([M − H]−) indicating an elemental composition of C15H14O6, and the MS/MS spectrum (Fig. 3E) is presented that the main fragment ions are m/z 245.08235, 203.07153. Among them, m/z 245.08235 is presumed by losing CO2 (44 Da) from [M − H]−, and m/z 203.07153 is inferred by losing of C2H2O (42 Da) from m/z 245.08235. The characterization of catechins in SHD is concluded by referring to the cleavage rule and related literatures.[21-23]

Characterization of lignans in SHD

Most of the lignans have the presence of allyl and hydroxyl groups, which relative positions are different, their cleavage methods will be different. Honokiol and magnolol were selected to illustrate the fragmentation pathways. Magnolol showed m/z 265.12394 ([M − H]−) in the negative ion mode, which indicated the elemental composition of C18H18O2. The mass spectrum (Fig. 3F) shows that the main fragment ions are m/z 247.10096, 245.09517, 223.07491. Among them, m/z 247.10096 is presumed by losing H2O (18 Da) from [M − H]−, m/z 245.09517 is inferred by losing H2 (2 Da) from m/z 247.10096, and m/z 223.07491 is obtained by losing C3H6 (42 Da) from [M − H]−. Honokiol showed m/z 265.12413 ([M − H]−) in the negative ion mode, which indicated the elemental composition of C18H18O2. The mass spectrum (Fig. 3G) shows that the main fragment ions are m/z 224.08126, 223.07514, 197.05923. Among them, m/z 224.08126 is presumed by losing C3H5 (41 Da) from [M − H]−, m/z 223.07514 is inferred by losing C3H6 (42 Da) from [M − H]−, and m/z 197.05923 is obtained by losing of C2H2 (26 Da) by m/z 223.07514. The characterization of lignans in SHD is concluded by referring to the cleavage rule and related literatures.[24,25]

Characterization of phenylpropanoid glycosides in SHD

The phenylpropanoid glycosides are composed of glucose and phenethyl alcohol, which mainly based on the loss of coffee groups (162 Da). Verbascoside is chosen as an example to explain the cleavage pathway. It showed m/z 623.19814 ([M − H]−) in the negative ion, which indicated an elemental composition of C29H36O15. The mass spectrum (Fig. 3H) shows that its main fragment ion is m/z 461.16314, 161.02265. Among them, m/z 461.16314 is presumed by losing C9H6O3 (162 Da) from [M − H]−, and m/z 161.02265 is inferred by losing C20H30O12 (462 Da) from [M − H]−. The characterization of phenylpropanoid glycosides in SHD is concluded by referring to the cleavage rule and related literatures.[24,25]

Characterization of alkaloids in SHD

The main rupture method of alkaloids is the loss of N. (R)-Oblongine is chosen as an example to explain the cleavage pathway. It showed an m/z 314.17473 ([M + H]+) in the positive ion mode, which indicated the elemental composition of C19H24NO3+. The mass spectrum (Fig. 3I) shows that the main fragment ions are m/z 269.11625, 237.09024, 175.07485, 107.04861. Among them, m/z 269.11625 is presumed by losing C2H6NH (45 Da) from [M + H]+, m/z 237.09024 is inferred by losing CH4O (32 Da) at m/z 269.11625, m/z 175.07485 is obtained by losing C6H6O (94 Da) from m/z 269.11625, and m/z 107.04861 is speculated by losing C12H18NO2 (207 Da) from [M + H]+. The characterization of alkaloids in SHD is concluded by referring to the cleavage rule and related literature.[25]

Characterization of coumarins in SHD

The main cleavage mode of coumarins is loss of CO and CO2, while coumarin glycosides lose glucose firstly, and then break by the above-mentioned cleavage law. Bergaptol is chosen as an example to explain the cleavage pathway. It showed an m/z 201.01957 ([M − H]−) in the negative ion mode, which indicated an elemental composition of C11H6O4. The mass spectrum (Fig. 3J) shows that the main fragment ions are m/z 185.02456, 157.17256, 129.10451, 101.96320. Among them, m/z 185.02456 is presumed by losing OH (17 Da) from [M − H]−, m/z 157.17256 is presumed by losing CO (28 Da) from m/z 185.02456, m/z 129.10451 is inferred by losing CO (28 Da) from m/z 157.17256, and m/z 101.96320 is obtained by losing CO (28 Da) from m/z 129.10451. The characterization of coumarins in SHD is concluded by referring to the cleavage rule and related literatures.[26-28]

Conclusions

In our study, a rapid, swift and straightforward analytical method with the help of UHPLC-FT-ICR-MS/MS was successfully developed for the first time to separate and identify the chemical constituents of SHD. A total of 137 compounds in SHD were identified or tentatively characterized, and 11 compounds were accurately identified by reference. The research indicated that SHD mainly contains 42 flavonoids (in Aurantii fructus immaturus and Rheum palmatum L.), 3 triterpenoids (in Aurantii fructus immaturus), 23 anthraquinones (in Rheum palmatum L.), 14 glucosides, 5 catechins and 2 anthranone (in Rheum palmatum L.), 11 phenylpropanoid glycosides, 5 alkaloids and 11 lignans (in Magnoliae Officmalis Cortex), 16 coumarins (in Notopterygii Rhizoma Et Radix and in Aurantii fructus immaturus),which has laid the foundation for prospective research associated with SHD. And it is hoped to be useful to identify and characterize the components in other TCMs. And it provided a basis for finding the prototype components and metabolites in the serum of containing SHD.

In this study, the structure of the compound, except for the comparison of the reference, was obtained through the combination of retention time, accurate primary and secondary mass spectrometry information, cracking rules of the same compound and literature comparison. The identification of compound structure by this method is limited to known compounds, because this method only mass spectrometry for identification.

Traditional Chinese medicines

Sanhua decoction

Aurantii fructus immaturus

Rheum palmatum L

Magnoliae Officmalis Cortex

Notopterygii Rhizoma Et Radix

Brain blood barrier

Ultra high performance liquid chromatography coupled with Fourier transform ion cyclotron resonance mass spectrometry

Base peak ion chromatograms

Extract ion chromatograms

Diels–Alder reaction

Conflicts of interest

The authors declare that there are no conflicts of interest.