• Users Online: 1972
  • Print this page
  • Email this page

Table of Contents
Year : 2021  |  Volume : 7  |  Issue : 3  |  Page : 347-360

Screening safflower injection for constituents with activity against stroke using comprehensive chemical profiling coupled with network pharmacology

State Key Laboratory of Natural Medicines, School of Traditional Chinese Pharmacy, China Pharmaceutical University, Nanjing, China

Date of Submission07-Dec-2020
Date of Decision03-Jun-2021
Date of Acceptance18-Feb-2021
Date of Web Publication9-Aug-2021

Correspondence Address:
Prof. Ping Li
State Key Laboratory of Natural Medicines, School of Traditional Chinese Pharmacy, China Pharmaceutical University, No. 24 Tongjia Lane, Nanjing 210009
Dr. Wen Gao
State Key Laboratory of Natural Medicines, School of Traditional Chinese Pharmacy, China Pharmaceutical University, No. 24 Tongjia Lane, Nanjing 210009
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/wjtcm.wjtcm_32_21

Rights and Permissions

Objective: This study aimed to explore safflower injection (SI) for constituents with activity against ischemic stroke using a combination of chemical analysis and a network pharmacology strategy. Materials and Methods: The main ingredients of SI were comprehensively identified using ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry, and the core targets and pathways associated with stroke were predicted using PharmMapper and Kyoto Encyclopedia of Genes and Genomes analysis. Cytoscape software was used to visualize and analyze the active compound-target-pathway network of SI regulating ischemic stroke. Results: A total of 76 chemical compounds were identified from the SI sample, including 63, which regulated 88 targets that were ultimately enriched in 12 key ischemia stroke-related signaling pathways. Kaempferol-3-O-sophoroside, kaempferol-3-O-rutinoside, carthamoside B6, neoeriocitrin, and 6-hydroxykaempferol-3-O-rutinoside-6-O-glucoside were determined to be important for stroke treatment because they had a higher degree value in the network than other constituents did. Moreover, the characteristic components isolated from SI showed protective effect mainly by acting on multiple targets including AKT1, epidermal growth factor receptor, transforming growth factor-beta receptor (TGFBR), Ras homolog, mTORC1 binding, caspase 3, and glycogen synthase kinase 3 beta, which were involved in different signaling pathways including phosphoinositide 3-kinase-Akt, mitogen-activated protein kinase, neurotrophin, ErbB, mechanistic target of rapamycin, and tumor necrosis factor. Conclusions: This study proposed a network pharmacology and chemical component profiling strategy for the systematic understanding of the therapeutic material basis of using SI against ischemic stroke.

Keywords: Chemical composition, ischemic stroke, network pharmacology, safflower injection, ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry

How to cite this article:
Shi XY, Miao QY, Liu XG, Li P, Gao W. Screening safflower injection for constituents with activity against stroke using comprehensive chemical profiling coupled with network pharmacology. World J Tradit Chin Med 2021;7:347-60

How to cite this URL:
Shi XY, Miao QY, Liu XG, Li P, Gao W. Screening safflower injection for constituents with activity against stroke using comprehensive chemical profiling coupled with network pharmacology. World J Tradit Chin Med [serial online] 2021 [cited 2022 Aug 8];7:347-60. Available from: https://www.wjtcm.net/text.asp?2021/7/3/347/323497

  Introduction Top

Traditional Chinese medicine (TCM) is a comprehensive therapeutic system for the prevention and treatment of various diseases, and the associated formulations are widely used for their mild healing effect and fewer side effects than conventional Western medicine.[1] Compared with agents used in Western medicine, TCM formulations generally contain hundreds of compounds and can exert their therapeutic effect in a holistic manner. The multicomponents of TCM formulations contribute to their synergetic effects on complex diseases.[2] However, the numerous and complex components are extremely difficult to analyze and screen for bioactive compounds, which is a major obstacle to the development of TCM formulations.[3] In addition, the ambiguity of active components also creates confusion in identifying the molecular action and therapeutic effectiveness, which seriously challenges the safety of TCM formulations.[4]

Safflower injection (SI) is a sterile water solution of the extract of flowers of Carthamus tinctorius L. (safflower). In China, SI has been officially approved as a clinical drug by the China Food and Drug Administration for the treatment of cerebral ischemia.[5] Sixteen compounds have been isolated from SI including safflower quinone glycoside, flavonoid, organic acid, and alkene compounds,[6] and 33 compounds including alkaloids and flavonoids were tentatively identified in normal and abnormal SI samples.[7] Furthermore, most studies on the pharmacological mechanism of SI have focused on the molecular action of hydroxysafflor yellow A (HYSA), a characteristic safflower uranidin, or other monomeric compounds.[8] At present, many studies have mainly focused on the qualitative and metabolic analysis of safflower,[9],[10] whereas few have conducted comprehensive chemical profiling of SI. Moreover, the molecular mechanisms of action of the multicomponents on multi-targets in the treatment of stroke have not been well elucidated.

In this study, an ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry (UPLC-QTOF/MS) method was developed for the comprehensive identification of the components of SI. Based on the optimized LC-MS conditions, an in-house chemical database including the name, molecular formula, molecular weight, precursor ion, and characteristic fragment ions of compounds reported from safflower was established, and their fragmentation pathways were derived. Then, the components of SI could be quickly identified based on the database search combined with the fragmentation pattern and comparison of reference standards. Furthermore, a network pharmacology analysis was conducted to predict the multi-target mechanism of action of SI against treating stroke.

  Materials and Methods Top

Chemicals and reagents

Acetonitrile and methanol (both are HPLC grade) were obtained from Merck (Darmstadt, Germany). HPLC grade formic acid and acetic acid were purchased from ROE Scientific Inc. (New Castle, USA). Deionized water was prepared using a Milli-Q water system (Millipore, MA, USA). The following reference standards HYSA (22), rutin (59), scutellarin (61), myristicin (24), luteolin-7-O-glucoside (65), kaempferol-3-O-rutinoside (70), quercetin-7-O-glucoside (60), isorhamnetin-3-O-rutinoside (71), p-hydroxybenzoic acid (11), adenosine (5), protocatechuic acid (7), protocatechualdehyde (10), p-hydroxybenzaldehyde (16), neochlorogenic acid (15), chlorogenic acid (21), 6-hydroxykaempferol-3-O-rutinoside-6-O-glucoside (54), 6-hydroxykaempferol-3, 6, 7-tri-O-glucoside (30), syringin (23), caffeic acid (18), and p-coumaric acid (25) were purchased from MUST Bio-Technology Co., Ltd. (Chengdu, China).

Furthermore, riboflavin (38) was obtained from National Institutes for Food and Drug Control (Beijing, China). Guanosine (4) and uridine (2) were acquired from Weikeqi Bio-Technology Co., Ltd. (Sichuan, China). L-pyroglutamic acid (1), D-phenylalanine (3), thymidine (8), L-tryptophan (9), roseoside (27), eriocitrin (57), neoeriocitrin (67), 6-hydroxykaempferol-3,6-di-O-glucoside (49), and kaempferol-3-O-sophoroside (52) were purchased from Yuanye Bio-Technology Co., Ltd.(Shanghai, China). The purities of all reference compounds were >98% except for 6-hydroxykaempferol-3,6-di-O-glucoside, which was >95%. SI was purchased from Ya'an Sanjiu Pharmaceutical Co., Ltd. (Sichuan, China).

Chromatographic and mass spectrometric conditions

A Waters ACQUITY UPLC system (Waters, Milford, USA) equipped with a binary pump, online degasser, autosampler, and column oven was used for sample analysis. Separation was performed using a Waters ACQUITY UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm), and an online filter (4.6 mm, 0.2 μm) was fitted before the column.

The column temperature was set to 30°C. The mobile phase consisted of 0.02% formic acid water (A) and acetonitrile (B) was used for gradient elution that was conducted as follows: 0–10 min, 0% B; 10–40 min, 0%–7% B; 40–55 min, 7%–7.5% B; 55–72 min, 7.5%–10% B; 72–90 min, 10%–15% B; 90–93 min, 15%–20% B; 93–110 min, 20%–31% B; 110–115 min, 31%–95% B; 115–118 min, 95% B; 118–119 min, 95%-0% B; and 119–123 min, 0% B. The injection volume and flow rate were 1.0 μL and 0.4 mL/min, respectively.

The H-Class Xevo G2-XS QTOF mass spectrometer (Waters, Milford, USA) with an electrospray ion source was operated in both positive and negative scanning modes. The ion source parameters were set as follows: capillary voltage, 3 kV; sample cone, 40 V; ion energy, 1.0 V; source and desolvation temperatures, 100°C and 400°C, respectively; and desolvation and cone gas flow, 600 L/h and 50 L/h, respectively.

When the collision energy (CE) was 6 eV, the scanning time was 0.5 s and the interval scanning time was 0.02 s. The data-dependent acquisition mode was adopted, and the scanning range was m/z 50–1200 Da. The lower limit and upper limit of CE were set at 5–10 V and 20–80 V, respectively. Data acquisition and processing were conducted using MassLynx 4.1 software (Waters, Milford, USA).

Preparation of standard solutions and samples

The 32 reference compounds were accurately weighed and dissolved in methanol to prepare 1 mg/mL standard stock solutions. Then, the 32 standard stock solutions were mixed, further diluted with methanol to a suitable concentration for analysis, and then centrifuged at 13,000 rpm for 10 min. The SI sample was directly centrifuged at 13,000 rpm for 10 min, and the supernatant solution was injected for LC-MS analysis.

Target prediction and network construction

Structures of 63 identified compounds were drawn using Chem3D and saved in the mol2 format. Furthermore, the PharmMapper Server (http://lilab-ecust.cn/pharmmapper/), which is a freely accessible web server that uses a pharmacophore mapping approach to predict drug targets,[11] was used to predict the targets of the ingredients of SI. Targets with a high fit score (Z-score >0.5) were screened to further improve the prediction confidence.

All target names were standardized using UniProt (http://www.uniprot.org/), and the target genes were imported into the online tool Database for Annotation, Visualization, and Integrated Discovery (DAVID) 6.8 (https://david.ncifcrf.gov) to perform Kyoto Encyclopedia of Genes and Genomes pathway enrichment. Pathways with P ≤ 0.05 were defined as significant, and among them, the ischemic stroke-related pathways were further selected as “key pathways” of SI.

Based on the targets in the above key pathways, the protein-protein interactions were obtained using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database (https://string-db.org/). Finally, the component-target-pathway (C-T-P) network of SI regulating ischemic stroke was constructed using Cytoscape 3.6.1 software. Three topological parameters including “degree,” “betweenness,” and “closeness” were calculated to assess the importance of nodes in the network.

  Results and Discussion Top

Optimization of ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry analysis conditions

The HPLC method was optimized to analyze the SI samples. The following different chromatographic parameters, chromatographic columns, mobile phase composition, elution gradient, column temperature, and flow rate were investigated to obtain a good peak shape and resolution. SI consists of many highly polar compounds with similar structure, and the Waters ACQUITY UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm) was finally chosen for the analysis because it showed better separation than the other columns did.

Compared with methanol-water, acetonitrile-water achieved satisfactory separation. Moreover, the addition of 0.02% formic acid improved the peak shape and ionization efficiency of the electrospray ion source. In this study, we also further examined the flow rate and column temperature, and the results showed that the best separation was achieved at 0.4 mL/min and 30°C, respectively. Finally, the optimized chromatographic conditions were determined as described in the section on chromatographic conditions.

For the MS conditions, the scanning mode and CE were optimized for the acquisition of the characteristic and effective fragments. The results showed that amino acids and flavonoids exhibited higher responses in the positive than in the negative mode, whereas organic acids and quinochalcones were more sensitive in the negative ion mode. In addition, we found that flavonoid O-glycosides produced abundant deglycation fragment ions at a low CE, but the quinochalcones only produced characteristic fragment ions under high CE conditions. In summary, both positive and negative ion modes were used for the SI analysis, and the lower and upper limits of the CE were set at 5–10 V and 20–80 V, respectively, which produced effective characteristic fragment ions for the comprehensive analysis of the fragmentation patterns of the compounds.

Comprehensive chemical profiling of safflower injection

Based on the optimized UPLC-QTOF/MS conditions described in the previous section, the comprehensive chemical profiling of SI was achieved, and the base peak intensity (BPI) chromatograms in the positive and negative ion modes are shown in [Figure 1]. To quickly identify the chemical composition of SI, the following information on the components previously isolated or identified from safflower was first retrieved from the literature to establish an in-house chemical database: the name, molecular formula, molecular weight, precursor ion, and characteristic fragment ions of compounds.
Figure 1: Negative (a) and positive (b) base peak intensity (BPI) chromatograms of safflower injection

Click here to view

The data acquisition and analysis of the SI sample were performed using a Waters MassLynx 4.1 workstation. The matched compounds were first confirmed using reference standards by comparison with their chromatographic and MS information under the same analysis conditions. The formula and fragment ions of compounds without standards or not included in the developed database were initially generated based on the accurate mass. Then, the structures of these compounds were deduced based on the diagnostic fragment ions and the fragmentation pathway according to the literature.[12],[13],[14],[15],[16],[17],[18],[19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31],[32],[33],[34],[35],[36],[37],[38],[39]

Combining with the commercial standard comparison, characteristic fragment analysis, and literature data, a total of 76 compounds were identified, including 15 quinochalcones, 22 flavonoids, 14 phenols (phenolic acids), 16 glycosides, 3 amino acids, 2 pyrimidines, 2 purines, 1 alkaloid, and 1 ether. The chemical structures of these compounds are shown in [Figure 2], and their relevant information such as retention time, molecular formula, adduct ion, precursor ion, mass deviation, and characteristic fragment ions are displayed in [Table 1].
Table 1: Characterization of 76 analytes of safflower injection

Click here to view
Figure 2: Chemical structures of the identified 76 compounds from safflower injection

Click here to view

Structural classification and identification

Structural characterization of C-glycosyl quinochalcones

C-glycosyl quinochalcones are unique constituents from safflower that have the same skeleton with that of the C-glycosylated cyclohexanonedienol moiety.[12] The structure generally contains a glucose residue linked with a C-C bond at the 5'-hydroxyl group, and different substituent groups, including pyrrole or furan ring, are commonly linked to the 3′-position.[13] In the tandem MS (MS/MS) spectra of quinochalcones, the neutral loss of C4H8O4 (120 Da) was usually obtained, and corresponded to the cleavage of the carbonyl group. In addition, consecutive losses of glucosyl were easily observed.[14]

Using the typical fragmentation pattern described above, 15 quinochalcones (12, 13, 19, 22, 31, 36, 37, 39, 40, 48, 50, 55, 64, 73, and 74) were identified from SI. Taking compound 22 as an example, a strong deprotonated molecular ion [M–H] was present at m/z 611.1624, and attributed to HYSA compared to the standards. A highly intense characteristic fragment ion at m/z 491.1183 [M–H–120] was produced by the loss of the C4H8O4 moiety. Other fragment ions were detected at m/z 473.1094 [M–H–C4H8O4–H2O], m/z 403.1041 [M–H–C6H10O5–H2O–CO], m/z 325.0723 [M–H–2C4H8O4–H2O–CO], and m/z 283.0608 [M–H–C4H8O4–C6H10O5–H2O–CO] through simultaneous or successive neutral loss of water (H2O, 18 Da), CO (28 Da), and glucose residue [162 Da, [Figure 3]a].
Figure 3: The cleavage characteristics of quinochalcones and the proposed fragmentation pathways of (a) hydroxysafflor yellow A; (b) saffloquinoside C; (c) tinctormine

Click here to view

The proposed fragmentation pathways of HYSA are shown in [Figure 3]. It is noteworthy that peaks 12, 13, and 19 showed identical [M–H] ions at m/z 611.16 at different retention times, and the same fragments were observed in the MS/MS spectra. All these compounds were identified as isomers of HYSA. Similarly, compound 40 produced an [M+H]+ ion at m/z 595.1633 with the elemental composition C27H30O15 in the positive ion mode. The secondary fragment ions showed the successive loss of glucose residue and H2O at m/z 433.1124 [M + H–C6H10O5]+, m/z 415.1021 [M+H–C6H10O5–H2O]+, which was tentatively identified as saffloquinoside C [Figure 3]b.[15] Compound 64 was inferred to be tinctormine [Figure 3]c with an [M–H] ion at m/z 592.1641 and fragment ions at m/z 472.1121 [M–H–C4H8O4].

Structural characterization of flavonoids

The main flavonoids identified from safflower were carthamidin, isocarthamidin, kaempferol, quercetin, and 6-hydroxykaempferol and its O-glycosides. In the negative ion mode, the fragmentation of most flavonoid O-glycogens involved the elimination of glycosyl residues and the acquisition of aglycone ions, i.e., m/z 285.04 (kaempferol), m/z 287.05 (carthamidin/isocarthamidin), and m/z 301.03 (quercetin/6-hydroxykaempferol).

The natural loss of the sugar molecule mainly included glucuronic acid (GlcA, C6H8O6, 176.03 Da), glucose (Glc, C6H10O5, 162.05 Da), and rhamnose (Rha, C6H10O4, 146.05 Da).[16] Taking compound 61 as an example, the [M–H] ion at m/z 461.0748 corresponded to the elemental composition C21H18O12. Furthermore, the fragment ion [M–H–C6H8O6] was present at m/z 285.0386 according to the loss of a GlcA residue. Compound 61 was identified as scutellarin through comparison with an authentic standard.

Compound 30 produced an [M–H] ion at m/z 787.1934 and characteristic product ions [M–H–C6H10O5], [M–H–2C6H10O5], and [M–H–3C6H10O5] at m/z 625.1445, m/z 463.0867, and m/z 301.0342, respectively. Consequently, it was plausibly identified as 6-hydroxykaempferol-3, 6, 7-tri-O-glucoside by further comparing its exact molecular masses, MS/MS spectral data, and retention times with those of authentic standards.

The fragmentation pattern of compound 59 was characteristic of a pseudomolecular ion [M–H] ion at m/z 609.1440, and the secondary product ions were at m/z 301.0378 [M–H–C6H10O4–C6H10O5] and m/z 284.0333 [M–H–C6H10O4–C6H10O5–OH], which was unequivocally assigned as rutin based on a comparison with the standard. Based on the proposed fragmentation pattern of flavonoids and commercial standards, 22 flavonoids were identified from SI, and were categorized as 8 flavone mono-O-glycosides (44, 58, 60, 61, 63, 65, 69, and 72), 12 flavone di-O-glycosides including quercetin-3,7-di-O-glucoside (34, 42, 45, 49, 52, 53, 56, 57, 59, 67, 70, and 71), and 2 flavone tri-O-glycosides (30 and 54).

Structural characterization of phenols (including phenolic acids)

The basic framework of phenols or phenolic acids consists of many hydroxyl, carbonyl, carboxyl, or aldehyde groups linked with phenyl groups. Owing to the existence of the above moieties, the loss of H2O (18 Da), CO (28 Da), and CO2 (44 Da) was easily observed in the MS/MS spectra of phenolic acids. In the present study, 14 phenols including phenolic acids (6, 7, 10, 11, 14, 15, 16, 17, 18, 20, 21, 25, 26, and 28) were identified from SI, and 8 were further confirmed through a comparison with authentic commercial standards. For example, p-hydroxybenzoic acid (compound 11) mainly yielded a deprotonated ion [M–H] at m/z 137.0245 and a fragment ion [M–H–CO2] at m/z 93.0329 by losing the carboxyl group. Compounds 15 and 21 were confirmed as neochlorogenic acid and chlorogenic acid, respectively, according to the reference standards. The compounds had the same elemental composition, C16H18O9, but different fragment ions. Neochlorogenic acid generated characteristic fragment ions at m/z 191.0546 [M–H–162 (C9H6O3)] and m/z 179.0350 [M–H–174 (C7H10O5)] by the loss of a quininic acid radical or caffeoyl residue, which also showed successive product ions at m/z 135.0443 and m/z 161.0235 through the loss of CO2 and H2O, respectively. However, chlorogenic acid yielded prominent fragment ions at m/z 325.0908 [M–H–CO] originating from the CO moiety, m/z 191.0546 and m/z 163.0360, by successive loss of a quininic acid radical and CO, respectively.

Structural characterization of glycosides

The glycosides isolated from SI generally exhibited a quasimolecular ion peak formed by [M+COOH] or [M+NH4]+ ions, and the neutral loss of the sugar moiety was easily obtained in the MS/MS fragment ions. For example, compound 23 showed a high abundance of the adduct ion [M+COOH] at m/z 417.1387, which was confirmed to be syringin by comparison with the reference standards. Its primary product ions appeared at m/z 357.1154 [M–H–CH2], m/z 209.0797 [M–H–C6H10O5], and m/z 194.0556 [M–H–C6H10O5–CH3]. In this study, 16 glycosides (23, 27, 29, 32, 33, 35, 41, 43, 46, 47, 51, 62, 66, 68, 75, and 76) were characterized in SI.

Structural characterization of amino acids

The amino acids showed a lower molecular weight in the prior retention time and responded both in positive and negative ion modes, which tended to lose 17 Da (–NH3), 44 Da (–CO2), 46 Da (–HCOOH or–H2O–CO), or a combination of these residues.[36],[37] Based on this observation, three amino acids, pyroglutamic acid (1), phenylalanine (3), and tryptophan (9), were ultimately identified by comparison with the reference standards and MS/MS fragment ion information. For example, compound 9 produced [M+H]+ at m/z 205.0963 and [M–H] ions at m/z 203.0822. The fragment ions of m/z 203.0822 were m/z 159.0932, 142.0636, and 116.0506, which were characterized as [M–H–CO2], [M–H–CO2–NH3], and [C8H6N], respectively.[38] In the positive ion mode, m/z 205.06 produced the characteristic fragment ions of tryptophan at m/z 188.0722 [M+H–NH3]+, m/z 159.0936 [M+H–H2O–CO]+, and m/z 146.0615.[39] Compared with the reference standard and the reported information,[38],[39] compound 9 was identified as tryptophan.

Structural characterization of other compounds

In addition to the above compounds, some other components were also identified in SI, including two pyrimidines (2 and 8), two purines (4 and 5), one alkaloid (38), and one ether (24). Compound 5 was identified as adenosine by comparison with authentic reference standards, which showed a predominant MS/MS fragment ion [M+H–Rib]+ at m/z 136.0621 through the loss of a ribose moiety (C5H8O4, 132.04 Da) in the positive ion mode. Compound 24 showed an adduct ion [M+H]+ at m/z 193.0863, and the successive fragment ions were characterized as m/z 161.0613 [M+H–OCH4], m/z 133.0641 [M+H–OCH4–C2H4], m/z 115.0542 [M+H–OCH4–C2H4–H2O], and m/z 105.0699 [M+H–OCH4–C2H4–CO], which was ultimately identified as myristicin.

Compound-target network construction and effective constituent screening of safflower injection for treating stoke

The components identified using UPLC-QTOF/MS were used to predict potential protein targets. Because some isomers could not be confirmed because of a lack of authentic reference standards, 63 chemical structures (one of the isomers was chosen as representative) were finally used for target prediction. The result showed that 411 potential targets were predicted based on the PharmMapper server, and they were further enriched into 114 significant signaling pathways.

Among these pathways, the following 12 [involving 88 targets, [Table 2] related to ischemic stroke were selected as the SI-regulated key pathways [Figure 4]: vascular endothelial growth factor/ErbB/neurotrophin/phosphoinositide 3-kinase (PI3K)-Akt/hypoxia-inducible factor-1/peroxisome proliferator-activated receptor/tumor necrosis factor (TNF)/mitogen-activated protein kinase/mechanistic target of rapamycin/Toll-like receptor/AMP-activated protein kinase/apoptosis signaling pathway. We discovered that up to 39 SI targets were enriched in the PI3K-Akt signaling pathway. Some studies have shown that the PI3K-Akt signaling pathway is closely related to proliferation, inflammation, and apoptosis, and its activation contributes to protecting against ischemia/reperfusion injury.[40]
Table 2: The targets enriched in the 12 signaling pathways related to ischemic stroke

Click here to view
Figure 4: The 12 key Kyoto Encyclopedia of Genes and Genomes pathway of safflower injection related to ischemic stroke (P≤0.05;theorderof importance was ranked from left to right by Log10 (P Value) with bar chart; the number of matching targets stick into each term with line chart)

Click here to view

To further interpret the complicated relationships mediating the SI-induced protection against ischemic stroke, a C-T-P network was constructed using Cytoscape [Figure 5]. The network displayed 163 nodes (63 compounds, 88 potential targets, and 12 pathways) and 2648 edges, including 1282 compound targets, 218 target pathways, and 1148 target-target interactions. The topological parameter analysis of all the nodes. The results showed that the mean degree value (target number associated with compound) was 20.35. Notably, kaempferol-3-O-sophoroside (C52, degree = 30), kaempferol-3-O-rutinoside (C70, degree = 30), carthamoside B6 (C28, degree = 29), neoeriocitrin (C67, degree = 28), and 6-hydroxykaempferol-3-O-rutinoside-6-O-glucoside (C54, degree = 28) were probably the potential active ingredients of SI mediating its activity against ischemic stroke because of their key positions in the network. Many targets such as AKT1 (degree = 94), epidermal growth factor receptor (EGFR, degree = 91), matrix metalloproteinase 9 (degree = 80), signal transducer and activator of transcription 1 (degree = 78), and glycogen synthase kinase 3 beta (GSK3B, degree = 72), were considered to be significant genes affected by multiple compounds or pathways. Recent studies have demonstrated that kaempferol-3-O-rutinoside, a primary flavonoid found in safflower, protects against cerebral ischemia by ameliorating oxidative stress and upregulating endothelial nitric oxide synthase activity.[41] Other common compounds with higher degree values have also been reported for the treatment of stroke. For example, rutin scavenges reactive oxygen species and attenuates ischemia-reperfusion injury.[42] Chlorogenic acid alleviates ischemic stroke by significantly reducing the expression of TNF, NOS2, and caspase 3 (CASP3).[39],[43] Furthermore, it is noteworthy that some characteristic ingredients of safflower such as C-glycosyl quinochalcones and kaempferol/6-hydroxykaempferol glycosides could also be regarded as the effective constituents of SI against ischemic stroke. These important compounds were found to be involved in the co-regulation of some key targets and pathways in the proposed C-T-P network. The combined overall effect of typical components of SI (C22, C28, C30, C33, C36, C40, C44, C49, C52, C53, C54, C64, C70, and C73) is shown in [Figure 6]. Some important targets such as TGFBR, EGFR, AKT1, GSK3B, Ras homolog, mTORC1 binding, and CASP3, were collectively hit by multiple compounds that acted on different ischemic stroke-related pathways to exert protective effects.
Figure 5: The ≴C-T-P” network of safflower injection on treating ischemic stroke (the compounds were represented by V-shape and their structure types were distinguished with different colors; the potential targets and pathways were shown by circles and triangles, respectively)

Click here to view
Figure 6: The proposed synergetic effect overview of characteristic components in safflower injection on treating ischemic stroke (the serial numbers represented typical compounds identified from safflower injection, and the targets regulated by various compounds were showed by different colors)

Click here to view

  Conclusions Top

In this study, an integrated UPLC-QTOF/MS and network pharmacology strategy was used to comprehensively analyze the components of SI with activity against ischemic stroke. In total, 76 compounds from SI were rapidly identified including quinochalcones, flavonoids, phenols (phenolic acids), glycosides, amino acids, and others. Among them, 32 components were identified by comparing them with reference standards. Furthermore, a pharmacology network analysis was used to determine the multiple mechanisms action of SI in treating ischemic stroke. The results demonstrated the neuroprotective effect of SI, which was probably mediated through its actions on 88 biological targets involved in 12 key signaling pathways related to stroke. The C-T-P network analysis revealed that kaempferol-3-O-sophoroside, kaempferol-3-O-rutinoside, carthamoside B6, neoeriocitrin, and 6-hydroxykaempferol-3-O-rutinoside-6-O-glucoside might be important bioactive components of SI. Moreover, the characteristic ingredients of safflower were also found to act on different targets and pathways related to ischemic stroke. In conclusion, the results presented in this study illustrate the viewpoint of the “multiple components-multiple targets-multiple pathways” of TCM system and contribute to elucidating the material basis of the effectiveness of SI.


This study was supported in part by the National Natural Science Foundation of China (Grant Nos. 81503241 and 81861168039).

Financial support and sponsorship


Conflicts of interest

Prof. Ping Li is an editorial Board member of World Journal of Traditional Chinese Medicine. The article was subject to the journal's standard procedures, with peer review handled independently of this editorial board member and their research groups.

  References Top

Jiang M, Zhang C, Cao H, Chan K, Lu SA. The role of Chinese medicine in the treatment of chronic diseases in China. Planta Med 2011;77:873-81.  Back to cited text no. 1
Li S, Hu YJ. Network pharmacology: An approach to the analysis of complex systems underlying traditional Chinese medicine. World J Tradit Chin Med 2018;4:135-6.  Back to cited text no. 2
  [Full text]  
Huang X, Kong L, Li X, Chen X, Guo M, Zou H. Strategy for analysis and screening of bioactive compounds in traditional Chinese medicines. J Chromatogr B 2004;812:71-84.  Back to cited text no. 3
Gao Y, Liang AH, Fan XH, Hu LM, Hao FR, Li YB. Safety research in traditional Chinese medicine: methods, applications, and outlook. Engineering 2019;5:76-82.  Back to cited text no. 4
Fan L, Pu R, Zhao HY, Liu X, Ma C, Wang BR, et al. Stability and degradation of hydroxysafflor yellow A and anhydrosafflor yellow B in the Safflower injection studied by HPLC-DAD-ESI-MSn. J Chin Pharm Sci 2011;20:47-56.  Back to cited text no. 5
Zhao JF, Liu J, Guo Y, Liu Q, Dai Z, Ma SC, et al. Chemical constituents from safflower injection and their bioactivity. China J Chin Mater Med 2014;39:3102-6.  Back to cited text no. 6
Feng WW, Zhang Y, Tang JF, Zhang CE, Dong Q, Li RY, et al. Combination of chemical fingerprinting with bioassay, a preferable approach for quality control of safflower injection. Anal Chim Acta 2018;1003:56-63.  Back to cited text no. 7
Sun Y, Xu DP, Qin Z, Wang PY, Hu BH, Yu JG, et al. Protective cerebrovascular effects of hydroxysafflor yellow A (HSYA) on ischemic stroke. Eur J Pharmacol 2018;818:604-9.  Back to cited text no. 8
Pu ZJ, Yue SJ, Zhou GS, Yan H, Shi XQ, Zhu ZH, et al. The comprehensive evaluation of safflowers in different producing areas by combined analysis of color, chemical compounds, and biological activity. Molecules 2019;24:3381.  Back to cited text no. 9
Chen JF, Song YL, Guo XY, Tu PF, Jiang Y. Characterization of the herb-derived components in rats following oral administration of Carthamus tinctorius extract by extracting diagnostic fragment ions (DFIs) in the MS(n) chromatograms. Analyst 2014;139:6474-85.  Back to cited text no. 10
Liu X, Ouyang S, Yu B, Liu Y, Huang K, Gong J, et al. PharmMapper server: A web server for potential drug target identification using pharmacophore mapping approach. Nucleic Acids Res 2010;38:W609-14.  Back to cited text no. 11
Qiang T, Liu J, Dong Y, Ma Y, Zhang B, Wei X, et al. Transcriptome sequencing and chemical analysis reveal the formation mechanism of white florets in Carthamus tinctorius L. Plants (Basel) 2020;9:847.  Back to cited text no. 12
Jin Y, Zhang XL, Shi H, Xiao YS, Ke YX, Xue XY, et al. Characterization of C-glycosyl quinochalcones in Carthamus tinctorius L. by ultraperformance liquid chromatography coupled with quadrupole-time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 2008;22:1275-87.  Back to cited text no. 13
Zhang QQ, Dong X, Liu XG, Gao W, Li P, Yang H. Rapid separation and identification of multiple constituents in Danhong injection by ultra-high performance liquid chromatography coupled to electrospray ionization quadrupole time-of-flight tandem mass spectrometry. Chin J Nat Med 2016;14:147-60.  Back to cited text no. 14
Jiang JS, Chen Z, Yang YN, Feng ZM, Zhang PC. Two new glycosides from the florets of Carthamus tinctorius. J Asian Nat Prod Res 2013;15:427-32.  Back to cited text no. 15
Jin Y, Xiao YS, Zhang FF, Xue XY, Xu Q, Liang XM. Systematic screening and characterization of flavonoid glycosides in Carthamus tinctorius L. by liquid chromatography/UV diode-array detection/electrospray ionization tandem mass spectrometry. J Pharm Biomed Anal 2008;46:418-30.  Back to cited text no. 16
Pu ZJ, Yue SJ, Yan H, Tang YP, Chen YY, Tan YJ, et al. Analysis and evaluation of nucleosides, nucleobases, and amino acids in safflower from different regions based on ultra high performance liquid chromatography coupled with triple-quadrupole linear ion-trap tandem mass spectrometry. J Sep Sci 2020;43:3170-82.  Back to cited text no. 17
Wang YQ, Tang X, Li JF, Wu YL, Sun YY, Fang MJ, et al. Development of an on-line mixed-mode gel liquid chromatography reversed phase liquid chromatography method for separation of water extract from Flos Carthami. J Chromatogr A 2017;1519:145-51.  Back to cited text no. 18
Zhou YZ, Chen H, Qiao L, Lu X, Hua HM, Pei YH. Five new aromatic glycosides from Carthamus tinctorius. Helv Chim Acta 2008;91:1277-85.  Back to cited text no. 19
Zhou YZ, Chen H, Qiao L, Xu N, Cao JQ, Pei YH. Two new compounds from Carthamus tinctorius. J Asian Nat Prod Res 2008;10:429-33.  Back to cited text no. 20
Joubert E, Beer DD, Malherbe CJ, Muller N, Bonnet SL, van der Westhuizen JH, et al. Occurrence and sensory perception of Z-2-(β-d-glucopyranosyloxy)-3-phenylpropenoic acid in rooibos (Aspalathus linearis). Food Chem 2013;136:1078-85.  Back to cited text no. 21
Olennikov DN, Kashchenko NI, Vennos C. A new esculetin glycoside from Calendula officinalis (Asteraceae) and its bioactivity. Farmacia 2017;65:698-702.  Back to cited text no. 22
Ren D, Ran L, Yang C, Xu M, Yi L. Integrated strategy for identifying minor components in complex samples combining mass defect, diagnostic ions and neutral loss information based on ultra-performance liquid chromatography-high resolution mass spectrometry platform: Folium Artemisiae Argyi as a case study. J Chromatogr A 2018;1550:35-44.  Back to cited text no. 23
Si W, Yang W, Guo D, Wu J, Zhang J, Qiu S, et al. Selective ion monitoring of quinochalcone C-glycoside markers for the simultaneous identification of Carthamus tinctorius L. in eleven Chinese patent medicines by UHPLC/QTOF MS. J Pharm Biomed Anal 2016;117:510-21.  Back to cited text no. 24
Chienthavorn O, Smith RM, Saha S, Wilson ID, Wright B, Taylor SD, et al. Superheated water chromatography-nuclear magnetic resonance spectroscopy and mass spectrometry of vitamins. J Pharm Biomed Anal 2004;36:477-82.  Back to cited text no. 25
Morikawa T, Wang LB, Ninomiya K, Nakamura S, Matsuda H, Muraoka O, et al. Medicinal flowers. XXX. Eight new glycosides, everlastosides F-M, from the flowers of Helichrysum arenarium. Chem Pharm Bull 2009;57:853-9.  Back to cited text no. 26
Zhou YZ, Qiao L, Chen H, Li RF, Hua HM, Pei YH. New aromatic glucosides from Carthamus tinctorius. J Asian Nat Prod Res 2008;10:817-21.  Back to cited text no. 27
Hong B, Wang Z, Xu T, Li C, Li W. Matrix solid-phase dispersion extraction followed by high performance liquid chromatography-diode array detection and ultra performance liquid chromatography-quadrupole-time of flight-mass spectrometer method for the determination of the main compounds from Carthamus tinctorius L. (Hong-hua). J Pharm Biomed Anal 2015;107:464-72.  Back to cited text no. 28
Yang X, Wang N, Shen C, Li H, Zhao J, Chen T, et al. An effective method based on medium-pressure liquid chromatography and recycling high-speed counter-current chromatography for enrichment and separation of three minor components with similar polarity from Dracocephalum tanguticum. J Sep Sci 2019;42:684-90.  Back to cited text no. 29
He J, Shen Y, Jiang JS, Yang YN, Feng ZM, Zhang PC, et al. New polyacetylene glucosides from the florets of Carthamus tinctorius and their weak anti-inflammatory activities. Carbohydr Res 2011;346:1903-8.  Back to cited text no. 30
Chiang YM, Chuang DY, Wang SY, Kuo YH, Tsai PW, Shyur LF. Metabolite profiling and chemopreventive bioactivity of plant extracts from Bidens pilosa. J Ethnopharmacol 2004;95:409-19.  Back to cited text no. 31
Dekany G, Wright K, Ward P, Toth I. Synthesis of sialyl lewis x analogues 2. J Carbohyd Chem 1997;16:11-24.  Back to cited text no. 32
He Y, Cheng P, Wang W, Yan S, Tang Q, Liu D, et al. Rapid investigation and screening of bioactive components in simo decoction via LC-Q-TOF-MS and UF-HPLC-MD methods. Molecules 2018;23:1792.  Back to cited text no. 33
Kurkin VA, Kharisova AV. Flavonoids of Carthamus tinctorius flowers. Chem Nat Compd 2014;50:446-8.  Back to cited text no. 34
Zhou YZ, Ma HY, Chen H, Qiao L, Yao Y, Cao JQ, et al. New acetylenic glucosides from Carthamus tinctorius. Chem Pharm Bull (Tokyo) 2006;54:1455-6.  Back to cited text no. 35
Kulik W, Heerma W. A study of the positive and negative ion fast atom bombardment mass spectra of α-amino acids. Biomed Environ Mass Spectrom 1988;15:419-27.  Back to cited text no. 36
Choi SS, Song MJ, Kim OB, Kim Y. Fragmentation patterns of protonated amino acids formed by atmospheric pressure chemical ionization. Rapid Commun Mass Spectrom 2013;27:143-51.  Back to cited text no. 37
Feketeová L, Khairallah GN, O'Hair RA, Nielsen SB. Gas-phase fragmentation of deprotonated tryptophan and its clusters [Trpn -H]- induced by different activation methods. Rapid Commun Mass Spectrom 2015;29:1395-402.  Back to cited text no. 38
Burrows EP. Dimethyl ether chemical ionization mass spectrometry of α-amino acids. J Mass Spectrom 1998;33:221-8.  Back to cited text no. 39
Luo S, Li H, Mo Z, Lei J, Zhu L, Huang Y, et al. Connectivity map identifies luteolin as a treatment option of ischemic stroke by inhibiting MMP9 and activation of the PI3K/Akt signaling pathway. Exp Mol Med 2019;51:1-1.  Back to cited text no. 40
Yu L, Chen C, Wang LF, Kuang X, Liu K, Zhang H, et al. Neuroprotective effect of kaempferol glycosides against brain injury and neuroinflammation by inhibiting the activation of NF-κB and STAT3 in transient focal stroke. PLoS One 2013;8:e55839.  Back to cited text no. 41
Ahmad H, Arya A, Agrawal S, Mall P, Samuel SS, Sharma K, et al. Rutin phospholipid complexes confer neuro-protection in ischemic-stroke rats. RSC ADV 2016;6:96445-54.  Back to cited text no. 42
Kumar G, Mukherjee S, Paliwal P, Singh SS, Birla H, Singh SP, et al. Neuroprotective effect of chlorogenic acid in global cerebral ischemia-reperfusion rat model. Naunyn Schmiedebergs Arch Pharmacol 2019;392:1293-309.  Back to cited text no. 43


  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]

  [Table 1], [Table 2]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Materials and Me...
Results and Disc...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded73    
    Comments [Add]    

Recommend this journal