2.1.1 Chemicals and reagents

A total of 8 reference standards were purchased from the National Institutes for Food and Drug Control (Beijing, China). They were 5-O-methylvisammioside (111523–201610) with the degree purity is 96.10%, Paeoniflorin (110736–201842) with the degree purity is 97.40%, Mangiferin (111607–201704) with the degree purity is 98.10%, Prim-O-glucosylcimifugin (111522–201712) with the degree purity is 96.20%, Liquiritin (111610–201607) with the degree purity is 93.10%, Gallic acid (110831–201605) with the degree purity is 90.80%, Ephedrine Hydrochloride (171241–201809) and Pseudoephedrine Hydrochloride (171237–201510).

Two reagents, acetonitrile and formic acid, met for HPLC grade. Both were bought from Thermo Fisher Scientific (USA). The Watsons distilled water was applied to preparation for samples and the mobile phase. All other reagents were just met for analytical grade.

2.1.2 Preparation of standard solutions

The eight reference standards mentioned above were dissolved in methanol to 10 ml to acquire 1 μg/ml of each reference substance, kept at 4 °C before analysis.

2.1.3 Preparation of sample solution

The material standard of GSZD according to the best decoction conditions obtained by Yi (2019). At first, Preparation of freeze-dried powder. Weigh Cinnamomum cassia Presl (also called Gui Zhi) 12 g, Paeonia lactiflora Pall. (Shao Yao) 9 g, Anemarrhena asphodeloides Bge. (Zhi Mu) 12 g, Saposhnikovia divaricate (Turcz.) Schischk. (Fang Feng)12 g, Atractylodes Macrocephala Koidz. (Bai Zhu) 15 g, Ephedrae sinica Stapf (Ma Huang) 6 g, Glycyrrhiza uralensis Fisch. (Gan Cao) 6 g, Zingiber officinale Rosc. (Sheng Jiang) 15 g, and Processed Aconitum carmichaelii Debx. (Pao Fuzi) 6 g respectively, added 1400 ml of water, soaked for 30 min, closed the lid of the casserole, boiled over high heat until boiling, opened the lid of the casserole, turned down the firepower to keep the liquid slightly boiling, the boiling time is about 70 min, filtered with 200 mesh filter cloth while hot and squeeze the dregs to cool, then fixed the volume to 400 ml, used vacuum freeze-drying to make freeze-dried powder. Second, accurately weighed 0.5 g of freeze-dried powder, added 25 ml of 75% methanol into a 50 ml Erlenmeyer flask respectively, next weighed by balance scales, then made use of the ultrasonic bath (40 kHz, 150 W) for 30 min, cooled to room temperature, adjusted the weight with methanol again, shaken, finally filtered sample solution with a 0.22 μm microporous membrane before analysis. The sample solution was stored at 4 °C before analysis.

2.1.4 Instrument and analysis conditions

Identification of GSZD compounds was carried out on the Vanquish Ultra High Performance Liquid Chromatography Q Exactive IV Pole-electrostatic field orbital trap high resolution mass spectrometer produced by Thermo Fisher Scientific (USA). Xcalibur software (version 3.0) was applied to command instruments, gather and analyze data.

Thermo Scientific Accucore™ C₁₈ column (100 × 3 mm, 2.6 μm, Thermo Scientific) was employed at column temperature 30 °C. The mobile phase consisted of 0.1% formic acid in water (A) and acetonitrile (B), and the flow rate was 0.3 ml/min. The gradient elution parameters was set as follows: 0–20 min, 5–70%B; 20–25 min, 70–90%B; 25–30 min, 90–5%B; 30–35 min, 5%B. The injection volume was 2 μL.

The positive and negative ion mode detection was performed on Q Exactive Quadrupole-Electrostatic Field Orbital Trap combined with UHPLC through ESI electrospray ion source. The spray voltage of positive mode and negative mode were both set at 3000 V, the probe heater temperature of positive mode and negative mode were both set at 350 °C, the sheath gas of positive mode and negative mode were both set at 35arb, the auxiliary gas of positive mode and negative mode were both set at 10arb, and capillary temperature of positive mode and negative mode were both set at 320 °C. S-Lens RF level was set at 50. The primary resolution of the full scan was 70,000 with the scanning range from m/z = 100 to m/z = 1500. The secondary resolution was 17,500 with the collision energy gradient of 20ev, 40ev, and 60ev.

2.1.5 Data processing

The raw data requirement and processing were carried out by Compound Discoverer 3.0 software, and On the condition of its wizard settings and method templates built the identification process of unknown compounds. Software executed peak alignment and peak extraction on the original data to obtain molecular ion peaks, and possible molecular formulas fitted by isotope. Moreover, the secondary fragments’ measured spectra were matched with databases named the mzCloud, the local Chinese medicine component database OTCML respectively. Preserve the chemical constituents met to peak area threshold is more than 80,000, the mass deviation is less than 5 ppm, and the matching score is more than 85. Finally, further confirming the compound with the reference substance and related literature.

2.2.1 Drug-likeness filtering

The identified chemical compounds were imported into the PubChem (https://pubchem.ncbi.nlm.nih.gov/) to acquire Canonical SMILES. The SwissADME (http://www.swissadme.ch/) base Canonical SMILES was utilized to identify “Drug-likeness” property according to the Lipinski’s rule and Gl absorption. The chemical components, meet condition of high Gl absorption, and more than two yes under “Druglikeness”, were filtered out as potential active ingredients.

2.2.2 Targets related to selected constituents or GA

For the putative targets genes of the selected compounds, two approaches were applied. First, the validated targets of potential active compounds were gathered from HERB (http://herb.ac.cn/), CTD (http://ctdbase.org/) with the species is “Homo sapiens”. Second, the possible targets were predicted by SEA (http://sea.bkslab.org/) under the condition with the screening criteria are: Tc＞0.7, and the species is “Homo sapiens”.

For the GA-related genes, they were searched with the phrase “gouty arthritis” as key word in the five public databases include CTD (http://ctdbase.org/), GeneCards (https://www.genecards.org/), OMIM (https://www.omim.org/), DisGeNET (https://www.disgenet.org/), GEO (https://www.ncbi.nlm.nih.gov/). All the GA-related targets were collected.

All targets, whether compound targets or disease targets, deleted duplicate values and invalid genes, and then all enter the Uniprot database (https://www.uniprot.org/) to calibrate the target protein and gene information. At last, they were all reviewed human targets.

The Venny 2.1 online tool (https://bioinfogp.cnb.csic.es/tools/venny/index.html) was exploited to draw the Venn diagram of active ingredient targets and GA targets, the existence of overlapping common targets may be the critical target of GSZD in the treatment of GA.

2.2.3 Screening and confirming active components and potential targets

The active ingredients and GA-related target interaction network was built by importing into Cytoscape (Version 3.8.0, Java 11.0.6) for visualization and analysis. The topological properties of the network were calculated using “Network Analyzer”, sort them according to the degree, Betweenness Centrality and Closeness Centrality values, take all the targets higher than the median, take the intersection as core targets. Analyze protein–protein interaction (PPI) of core targets in String database (https://string-db.org/), where the species were limited to “Homo sapiens”, and set the minimum required connection score between the targets is 0.4, and import Cytoscape software for processing and screening.

2.2.4 Enrichment analysis

The ClueGo plug-in was often used for Go enrichment analysis. At first, inputting the selected key targets for biological processes, molecular functions, cellular component, and KEGG pathway enrichment analysis respectively. Second, select “Homo sapiens”. The set the P and advanced term/pathway selection options. Besides, the REACTOME database (https://reactome.org/) was utilized for pathway enrichment and screening. Based on the relationship between GSZD and core active components, and the relationship between core targets and pathway, GSZD-herb-core active component-target-pathway was visualized by Cytoscape 3.8.0 software.

3.1.1 Identification of chromone compounds

Chromone compounds are unique to Saposhnikovia divaricate (Turcz.) Schischk., which was demonstrated to inhibit arthritis (Kong et al., 2013). It is easy to make a conclusion that the class of components generally displayed [M+H]⁺ under the first-level mass spectrum. And under the secondary mass spectrometry, it mainly cracks in accordance with Retro-Diels-Alder (RDA) rule or loss of neutral fragment ions contained glucoside (1 6 2), methyl (15), H₂O (18), CO (28), CO₂ (44). It was just identified 3 chromone compounds from GSZD with databases, related literature, and reference standards. Compound 65 exhibited a [M+H]⁺ which the quasi-molecular ion at m/z = 453.17575, and the molecular formula was conjectured to C₂₂H₂₈O₁₀. In the MS/MS spectrometry, at first, it became m/z = 291.12265 [M+H-C₆H₁₀O₅]⁺ by losing glucose, and then separately decomposed into four fragment products of m/z = 273.11206 [M+H-C₆H₁₀O₅-H₂O]⁺, m/z = 243.06525 [M+H-C₆H₁₀O₅-CH₄O₂]⁺, m/z = 231.06529 [M+H-C₆H₁₀O₅-C₃H₈O]⁺, m/z = 219.06546 [M+H-C₆H₁₀O₅-C₄H₈O]⁺ by losing some neutral fragment ions such as C₃H₈O, H₂O, C₄H₈0 or generating the RDA cleavage. M/Z = 216.04190 [M+H-C₆H₁₀O₅-C₃H₈O-CH₃]⁺, m/z = 203.07022 [M+H-C₆H₁₀O₅-C₃H₈O- CO]⁺ fragment ion may be required by losing CH₃, CO under m/z = 231.06529 fragment ion. In accordance with the database and the reference standard, compound 65 was determined to 5-O-methylvisammioside (Chen et al., 2018; Liu et al., 2018; Shao et al., 2018). The possible MS fragmentation pathway of 5-O-methylvisammioside is displayed in Fig. 3 A.

Close

Fig. 3

The possible fragmentation pathway of chemical. components. A shows compound 65; B shows compound 56 and.

Export to PPT

3.1.2 Identification of flavonoids compounds

Flavonoids are comprehensively distributed in Chinese medicinal herbs with enormous pharmacological activities like neuroprotective, anti-inflammatory, anti-oxidant, and anti-tumor after referring to related literature (Duan et al., 2019; Piwowar et al., 2020; Zeng et al., 2020; Zhai et al., 2019). They are often conjugated with pentose, Hexose, and hexuronic acid. In accordance with the structure, they are roughly compartmentalized into flavonoids, flavonols, dihydroflavonoids, isoflavones, chalcones, etc. These compounds have roughly similarity in cleavage patterns, except for isoflavones. Under the circumstance of first-level mass spectrometry, these compounds can be monitored in both positive and negative ion modes. Furthermore, the characteristic fragments of glycosides were generally lost carbohydrate structure contained pentose (1 3 2), Hexose (1 6 2), hexuronic acid (1 7 6), as well as neutral molecular fragments contained CHO (15), H₂O (18), CO (28), CO₂ (44). Of course, RDA cleavage reaction also arose in the MS². It was totally identified 26 flavonoids from GSZD, principally from three herbs (Anemarrhena asphodeloides Bge., Ephedrae sinica Stapf, Glycyrrhiza uralensis Fisch.). For instance, compound 56 displayed a [M−H]⁻ which the quasi-molecular ion at m/z 417.11996, and the molecular formula was conjectured to C₂₁H₂₂O₉. In the MS/MS spectrum, after losing one molecule of glucose, the component became m/z = 255.0664[M-H-C₆H₁₀O₅]⁻ which then decomposed into fragment product ions contained m/z = 153.01900, m/z = 135.00833, m/z = 119.04967, m/z = 91.1820 due to losing some neutral fragment ions include C₃H₈O, H₂O, C₄H₈O or generating the RDA cleavage reaction. By comparison with chemical databases, literature as well as reference substance (Li, 2015; Zhou et al., 2004), which all demonstrated that compound 56 could be deemed to Liquiritin. Compound 71 bore a resemblance to the cleavage of compound 56, which they both possessed 417, 255, 153, 135, 119, and 91 ion fragments. Moreover, it displayed a [M−H]⁻ which the quasi-molecular ion at m/z 549.16272 as well as the predicted molecular formula was C₂₆H₃₀O₁₃. As a consequence, the component was considered to Liguiritigenin-7-O-β-D-apiosyl-4′-O-β-D-glucoside, in accordance with the related literature (Yang et al., 2020b; Zhou et al., 2004). The possible MS fragmentation pathway of component 56 and 71 is displayed in Fig. 3B.

3.1.3 Identification of alkaloid compounds

Alkaloid compounds are comprehensively distributed in Chinese medicinal materials with enormously biological activity, whereas the majority of them may result in inevitably adverse effects. Such compounds manifested significant activities in inflammatory and tumor disease referred to research articles (Chen et al., 2020; Tong et al., 2013; Zheng et al., 2013). The class of components generally exhibited [M+H]⁺ under the conditions of first-level mass spectrometry. Losing the substituents in the MS/MS spectrum is a hallmark of a class of components, such as hydroxyl, amino, methoxy, acetyl, benzoyl, etc. Another identifying feature of the class of components is losing neutral fragment molecules that contain CO(28), H₂O(18), NH₂(16), CH₃OH(32), CH₃COOH(60), C₇H₆O₂ (1 2 2), etc. It was totally identified 10 alkaloid constituents from GSZD, principally from three herbs (Ephedrae sinica Stapf, Processed Aconitum carmichaelii Debx.,and Anemarrhena asphodeloides Bge.). Compound 26 displayed a [M+H]⁺ which the quasi-molecular ion at m/z 166.12265, and the molecular formula was conjectured to C₁₀H₁₅NO, after losing one molecule of H₂O, it became m/z = 148.11197[M+H-H₂O]⁺ which then decomposed into fragment product ions contained m/z = 133.08862, m/z = 117.07005, m/z = 91.05464 due to losing the methyl, NH_2, and vinyl. Component 26 was considered to Ephedrine by comparison with databases, related literature as well as reference standards (Meng et al., 2006). Component 29, which exhibited a [M+H]⁺ (m/z 166.12283), possessed predicted molecular formula (C₁₀H₁₅NO) identical to component 26 in the positive ion mode. Moreover, they both owned homogeneous product ion fragments at 148, 133, 117, and 91, It is not hard to infer that component 29 was Pseudoephedrine by comparison with reference standards. The possible MS/MS fragmentation pathway of Component 26 is manifested in Fig. 3C.

3.1.4 Identification of phenolic acid compounds

There are 10 phenolic acids confirmed from GSZD in total, which mainly from one herb named Paeonia lactiflora Pall. The class of components is generally concentrated in the negative ion mode. Furthermore, neutral fragment molecules contained CO (28), H₂O (18), CO₂ (44), C₇H₆O₂ (1 2 2) can be deemed to identifying features for major phenolic acids. For example, compound 31, which displayed quasi-molecular [M−H]⁻ ion at m/z 169.1385, and the molecular formula was conjectured to C₇ H₆ O₅, was decomposed into m/z = 125.0237[M-H-CO₂]⁻ after undergoing neutral loss of CO₂, then generated the fragmentation product of m/z = 97.02866, m/z = 81.03369, m/z = 69.03363 with losing CO and H₂O. In agreement with the standard reference, compound 31 was deemed to gallic acid. Compound 44, with quasi-molecular ion at m/z 353.08881 [M−H]⁻ as well as the molecular formula was conjectured to C₁₆ H₁₈ O₉, decomposed into four fragment products of m/z 191.05589, m/z 179.03462, m/z 161.02374, and m/z 135.04454. As a consequence, it was considered to Neochlorogenic acid while referring to the previous literature and database (Ouyang et al., 2017). The possible fragmentation pathway of compound 31 and 44 are revealed in Fig. 3D. the quasi-molecular ion and the MS² information of compound 43 were keeping with compound 44. Consistent with the mzCloud and mzVault database, and it was regarded as Chlorogenic acid potentially.

3.1.5 Identification of nucleosides and nucleobases compounds

Nucleosides and nucleobases are commonly discovered in almost every herb, whereas omitted by scholars. Anyway, they manifest widespread biological activities beneficial to human health. It was identified 7 components from GSZD totally. Loss of ribose and NH₃ can be deemed to a cleavage feature of nucleosides and nucleobases compounds. In compound 24, the quasimolecular [M+H]⁺ ion was at m/ z 268.104 and the molecular formula was conjectured to C₁₀H₁₃N₅O₄, only with a strong fragment ion at m/z 136.06178. Compound 24 was eventually deemed to be Adenosine, consistent with relevant literature (Yang et al., 2020a). The possible fragmentation pathway is exhibited in Fig. 3E.

3.1.6 Other types of compounds

Besides, sugars, amino acids, coumarins, vitamins, and terpenoids were identified from GSZD as well. carbohydrate constituents like D-Raffinose were decomposed into identical fragment ions of m/z = 59, m/z = 71, m/ z = 89, m/z = 179 on the condition of the negative ion mode. Moreover, such as compound 47, regarded as paeoniflorin in consistent with the reference substance that only discovered in Paeonia lactiflora Pall., possessed cleavage features mainly includes cracking glycosidic bonds, lossing sugar, benzoic acid as well as neutral molecules, result in forming characteristic fragment products with m/z = 121 and m/z = 121.