Studies on Quality Markers of Kaihoujian Spray for Anti-Inflammation Based on Gray Correlation Analysis Strategy

Kaihoujian spray (KHJ) was originated from the classical prescription of Miao medicine, which was commonly used for acute and chronic pharyngitis. The prescription was composed of Sophorae Tonkinensis Radix, Ardisiae Radix, Cicadae Periostracum, and menthol. However, in previous literature, only clinical studies have been reported. The Quality Marker (Q-Markers) of KHJ on anti-inflammation has not been clearly elucidated. In this study, a gray correlation analysis strategy combined with network pharmacology analysis was established for the investigation of Q-Markers in KHJ. A total of 52 components were identified or tentatively characterized in KHJ, including alkaloids, saponins, bergenin, flavonoids, amino acids, and their derivatives. Furthermore, regularity of recipe composition and gray correlation analysis revealed that the correlation degree of all peaks was greater than 0.5. The ranking of correlation degree was peak 1 > 6>9 > 8>7 > 10>4 > 5>11 > 3>2. Among them, peaks 2, 4, 5, 6, 8, 9, and 11 were identified as anagyrine, matrine, sophocarpine, norbergenin, bengenin, 11-O-galloylbergenin, and trifolirhizin. The network pharmacology analysis revealed that EGFR, MMP9, MMP3, MMP1, and PTGS2 were the main targets of KHJ. Bergenin, matrine, sophocarpine, calycosin, and trifolirhizin were the main anti-inflammatory active ingredient in KHJ. These results proposed that bergenin, sophocarpidine, sophocarpine, and trifolirhizin could be the Q-Markers of KHJ on anti-inflammation. The process of discovering the Q-Markers would provide a promising method of quality control on KHJ.


Introduction
As a significant portion of traditional Chinese medicine (TCM), Miao medicine has a long history of three or four thousand years. It is generally considered to be mysterious and magical and has its own system, especially famous for its external treatment of internal diseases [1]. Miao herb formulation (MHF) is a valuable medical experience accumulated by Miao folk in their long-term production activities and the practice of fighting against diseases and injuries.
ey have a profound understanding of etiology, elements, disease diagnosis, treatment, and prevention and have many unique features in clinical prescription and medication [2]. eir abundant medical experience has enriched the culture and become an important part of TCM.
However, similar to TCM, MHF also has many problems, such as unclear material basis and index components. Although some MHF were included in Chinese Pharmacopoeia, the quality standards were only established on the basis of their major components. Whether the index component was related to its efficacy was still dubious.
Fortunately, the concept of Quality Marker (Q-Marker) was established by Liu et al. [3] for the development and improvement of the quality of TCM. e candidates for Q-Markers should meet these criteria [3,4]: (1) e candidates should exist in original materials, TCM products, or formed during processing and preparation. (2) e candidates should be unique to some herbs and not derived from other herbs.
Kaihoujian spray (KHJ) was originated from the classical prescription of Miao medicine, which was a commonly used Chinese patent medicine for children with acute and chronic pharyngitis, and produced by Guizhou Sanli Pharmaceutical Limited by Share Ltd. e prescription was composed of Sophorae Tonkinensis Radix, Ardisiae Radix, Cicadae Periostracum, and menthol. According to the previous literature, Sophorae Tonkinensis Radix has the effects of antiinflammation [15], antivirus [16], inhibiting bacteria [17], improving immunity [18], and so on. Ardisiae Radix has the effects of bacteriostasis [19], analgesia [20], antivirus [21], and so on. Cicadae Periostracum has the effects of antiallergic [22], antitussive and antiasthmatic [23], bacteriostatic [24], and so on. Menthol has the effects of analgesic [25], osmotic [26], and so on. Kaihoujian spray can directly act on oral mucosa and avoid first-pass effect without gastrointestinal absorption and has the clinical advantages of fast onset, high bioavailability, small side effects, short course of treatment, convenient medication, and high patient compliance [27]. At present, the Q-Markers of KHJ on anti-inflammation have not been clearly elucidated. Only clinical studies have been reported in previous literature. Hence, it was necessary to develop a strategy to discover and validate the Q-Markers of KHJ on anti-inflammation. In this study, a gray correlation analysis strategy combined with network pharmacology analysis was established for the investigation of Q-Markers on KHJ. e results showed that bergenin, sophocarpidine, sophocarpine, and trifolirhizin should be the Q-Markers of KHJ on anti-inflammation. e process of discovering the Q-Markers would provide a promising method of quality control on KHJ.

Materials and Chemicals.
e Sophorae Tonkinensis Radix (the dried root of Sophora tonkinensis Gagnep.), Ardisiae Radix (the dried root of Ardisia crenata Sims.), Cicadae Periostracum (the dried shell of Cryptotympana pustulata fabricius), and menthol were provided by Guizhou Sanli Pharmaceutical Limited by Share Ltd. and identified by the Researcher Chengwang Tian. e voucher specimens (STR-2019, AR-2019, CPS-2019, and MEL-2019) were stored in herbaria at Tianjin Institute of Pharmaceutical Research, China.
e Griess reagents were purchased from Tianjin Guangfu Chemical Research Institute. Dulbecco's modified eagle's medium (DMEM) and Fetal Bovine Serum (FBS) were purchased from Gibco Ltd. All organic solvents used in this study were of HPLC grade and purchased from Concord Technology Co., Ltd. Pure distilled water was purchased from Wahaha Group Co., Ltd. (Hangzhou, China).

Preparation of Samples.
e whole prescription sample was prepared according to the method for KHJ Spray Standard published by National Medical Products Administration in 2002. In a brief description, Sophorae Tonkinensis Radix (250 g), Ardisiae Radix (250 g), and Cicadae Periostracum (250 g) were refluxed with pure water twice (1 : 10, w/v, 2 h each). After evaporation of the solvent in vacuo, ethanol was added to the residues until the ethanol content reached 80%. While kept standing for 24 hours, the solvent was filtered and evaporated in vacuo. e crude extract was mixed with menthol (1 g). Samples for regularity of recipe composition (KHJ-1∼14) were prepared as the above method, except for the assigned herbs. e ingredients of each sample are shown in Table 1.

HPLC/Q-Tof-MS/MS Analysis of KHJ.
e sample for HPLC/Q-TOF-MS/MS analysis was prepared as follows: approximately 0.1 g of the whole prescription sample (KHJ-15) was dissolved and diluted to 10 mL by 40% methanol. e sample was filtered through a 0.22 μm membrane filter and then injected 10 μL filtrate into the HPLC system for analysis. Chromatographic separation was carried out on an ultimate Plus C18 column (4.6 × 250 mm,5 μm). e mobile phase was optimized as 0.1% aqueous formic acid (A) and acetonitrile (B), and the gradient of elution was as follows: 0-15 min, 3%-8% B; 15-25 min, 8%-15% B; 25-35 min, 15%-22% B; 35-45 min, 22%-24% B; 45-75 min, 24-50% B. e flow rate was 1.0 mL/min, and the column temperature was held at 30°C. e optimum absorbed wavelength was selected as 210 nm according to the favorable resolution and multiple chromatographic peaks. e mass spectrometry analysis was obtained on a Sciex X500 R QTOF mass spectrometer equipped with an electrospray interface (ESI) source (AB Sciex, Framingham, MA, USA). Positive and negative ion modes were used for detection, the capillary voltages were 5500 V and 4500 V, and the cleavage voltage was 50 V and 80 V, respectively. e curtain gas was 35 PSI, and the atomizing temperature was 600°C. e mass data were achieved in the range of m/z from 50 to 1800 Da with a response value of more than 100 cps of the four highest peaks for secondary mass spectrum scanning. Data were collected and analyzed by analyst software SCIEX OS 1.4.

HPLC-DAD Analysis for Samples of Regularity of Recipe
Composition. e analysis of the regularity of recipe composition samples (KHJ-1∼14) was performed by the Waters e2695 (United States) with a PDA detector. e gradient of elution for mobile phase, optimum absorbed wavelength, flow rate, chromatography column, and column temperature were set as HPLC/Q-TOF-MS/MS method.

Inhibition of NO Production in LPS-Induced RAW264.7
Cells. RAW264.7 cells were incubated and divided into several groups. After stimulation with and without LPS (2 mg/mL) for 24 hours, the supernatant of media was collected for NO production analysis. 50 μL supernatant mixed with 50 μL Griess reagent was incubated in the dark for 10 min at 37°C. e OD value of each well was measured by a microplate reader at 540 nm. e concentration of NO was determined by the standard curve from sodium nitrite.

Statistical
Analysis. e measurement data were analyzed by IBM SPSS 23.0 (USA) and expressed as means ± SD (n � 3). e data satisfying normality and homogeneity of variance were analyzed by one-way ANOVA, and the comparison between groups was performed by the LSD method. If the data do not meet the homogeneity of variance, using K independent samples nonparametric test, P < 0.05 has statistical significance.

Network Pharmacology Analysis.
All the identified compounds in HPLC/Q-Tof-MS/MS analysis of KHJ were selected for the target compounds. en these target compounds were introduced into a SwissTargetPrediction database (https://www.swisstargetprediction.ch) to predict an action target of the active compound. e TCMSP database (https://tcmspw.com/tcmsp.php) was searched for the possible targets of the active ingredients of KHJ, and then the Dragbank database (https://www.drugbank.ca/) and Uniprot database (http://www.uniprot.org) were searched for the target GeneSymbol. Taking "inflammation" as the keyword, we searched the DigSee database (http://digsee. com) to obtain the targets related to anti-inflammatory and screened the inflammatory targets with a correlation degree greater than 0.8. e data of active compounds and inflammatory targets screened above were sorted and imported into Cytoscape 3.7.0 software to construct the active ingredient-disease target network, and the network topology was analyzed by Network Analyzer. e intersection targets of drugs and diseases were imported into the STRING database (https:// www.string-db.org), the confidence value was set to 0.4, the target interaction data were exported and processed by Excel and then imported into Cytoscape 3.7.0 to realize visualization, and the topology analysis of PPI network was carried out. rough the ClueGO, gene ontology (GO) analysis was carried out on the core target of KHJ under the conditions of number of genes � 3 and min percentage � 4.0. Homo sapiens, overlap >3, P < 0.01, and enrichment >1.5 were selected as screening conditions for GO and KEGG analysis. e pathway with lower P value and more enriched genes was screened, the GO analysis was drawn by R language ggplot2 software package, and the bubble map of KEGG analysis was drawn by Origin Pro 2021 software. Pathways with a smaller P value and more enriched genes were screened, the data of drug flavor, components, targets, and pathways were sorted out and introduced into Cytoscape 3.7.0 software to construct the network of drug flavorcomponent-target-pathway, and the network topology was analyzed.

Analysis of Chemical Ingredients in KHJ.
e total ion chromatography (TIC) of KHJ in positive and negative modes is shown in Figure 1. A total of 52 components were identified or tentatively characterized in KHJ, including alkaloids, saponins, bergenin, flavonoids, amino acids, and their derivatives. Among them, 32 compounds (alkaloids, flavonoids, saponins, etc.) were derived from Sophorae Tonkinensis Radix, 14 compounds (coumarins, saponins, etc.) were derived from Ardisiae Radix, and 6 compounds (amino acids) were derived from Cicadae Periostracum. e identification of these compounds was mainly based on the

Identification of Bergenin Derivatives in KHJ.
Bergenin and its derivatives are the main effective constituents of Ardisiae Radix and have the effect of relieving cough by inhibiting the cough center. In this study, seven bergenin and its derivatives from Ardisiae Radix were identified in KHJ.
Evidence-Based Complementary and Alternative Medicine   Figure 4.

Identification of Amino and Its Derivatives in KHJ.
Seven nitrogenous compounds from Cicadae Periostracum were identified from KHJ and speculated to be amino acids and acetyldopamine dimers. Amino acid was the main component, and acetyldopamine dimer was the main antiinflammatory and antioxidant component of Cicadae Periostracum [24]. Take   us, compound 39 was identified to be trifolirhizin. e mass spectrum and possible cleavage pathways of compound 39 are shown in Figure 6.

Identification of Saponin in KHJ.
Nine saponins were identified from KHJ. Among them, subprosides V, subproside II methyl ester, soyasaponin I, and kudzusaponin A3 were isolated from Sophorae Tonkinensis Radix, and ardisicrenoside B, ardisicrenoside H, ardisicrenoside G, ardisicrenoside N, and ardisiacrispin A were isolated from Ardisiae Radix. Take

Cytotoxicity of KHJ against RAW264.7 Cells.
e cytotoxicity of the regularity of recipe composition samples and the whole prescription sample was determined by MTT assay [30]. e results showed that KHJ had no cytotoxicity on RAW 264.7 cells at the doses of 1-100 μg/mL. e cell survival rate of KHJ-2, KHJ-4, and KHJ-9 was significantly lower than that of the blank group (P < 0.05) at the dose of 200 μg/mL, indicating that KHJ-2, KHJ-4, and KHJ-9 at the concentration of 200 μg/ mL could inhibit the proliferation of RAW264.7 cells.

Inhibition Effect of KHJ on NO Production in LPS-Induced
RAW 264.7 Cells. As shown in Figure 8, LPS-induced RAW264.7 cells can significantly promote the production of NO, and the content of NO in KHJ-4 had no significant difference compared with that in the model group (P > 0.05). Other groups could inhibit the release of NO in RAW264.7 cells to different degrees (P > 0.05).       Table 3, KHJ-2, KHJ-9, KHJ-5, KHJ-11, and KHJ-1 groups had a stronger anti-inflammatory effect in vitro. ese results indicated that the anti-inflammatory active ingredients of KHJ mainly came from Sophorae Tonkinensis Radix and Ardisiae Radix.

Gray Correlation Analysis.
Gray relational analysis (GRA) was a quantitative description and comparison method for the development and change of a system. Its basic idea was to judge whether the relationship was close by determining the geometric shape similarity between the reference data column and several comparative data columns, which reflects the correlation degree between           the curves. In this study, the spectrum-effect relationship of KHJ was studied through the GRA method. A total of 11 main peaks were selected according to the HPLC-DAD analysis for the regularity of recipe composition samples (Figure 9). e anti-inflammatory activity score of each sample was taken as a reference column, and the   correlation degree was calculated after the original data was treated with the dimensionless standard. e results of peak area and spectrum-effect correlation analysis are shown in Table 4. As shown in Table 4, the correlation degree of all peaks was greater than 0.5.
is indicated that ingredients in KHJ were acting in synergy. e ranking of correlation degree was peak 1 > 6>9 > 8>7 > 10>4 > 5>11 > 3>2. Among them, peaks 2, 4, 5, 6, 8, 9, and 11 were identified as anagyrine, matrine, sophocarpine, norbergenin, bengenin, 11-O-galloylbergenin, and trifolirhizin through HPLC/Q-Tof-MS/MS analysis. ese ingredients could be candidates for the Q-Markers of KHJ on antiinflammation. Figure 10, the active ingredient of the drug-potential target was visualized by Cytoscape 3.7.0. e key active ingredients, which are higher than the average value and the core target of KHJ, were imported into the STRING database and processed by Cytoscape 3.7.0 to obtain the PPI network ( Figure 11). e network consists of 65 nodes and 293 edges. e targets whose values were greater than the average degree value were as follows: GAPDH, EGFR, TNF, PTGS2, MMP 9, CCND1, ESR1, AR, PLAU, MMP3, AGTR1, MMP1, ADAM17, IL2, CTSB, MMP7, NOS2, PARP1, MME, and TLR9. Metascape database (https: metascape.org) was used for GO enrichment analysis and KEGG analysis of 20 key targets. e 20 key targets    Evidence-Based Complementary and Alternative Medicine were mainly involved in biological functions such as ultraviolet response, inflammation regulation, collagen catabolism process, light stimulation response, nerve inflammation response, and external stimulation response by influencing the activities of metalloendopeptidase, serine proteolytic enzyme, and serine-type endopeptidase.

Network Pharmacological Analysis. As shown in
e results are shown in Figure 12(a). e 20 key targets were analyzed for GO enrichment by ClueGO. As shown in Figure 12(b), 20 key targets of KHJ were mainly involved in the nitric oxide synthase activity signal transduction pathway, ultraviolet radiation response signal pathway, nerve inflammation response, and collagen catabolism process.
e Metascape database was used for KEGG analysis, involving 61 entries, in which the cancer pathway, prostate cancer pathway, and interleukin-17 signal pathway were enriched with more genes and smaller P value. ese results are shown in Figure 13. e data of compounds, targets, and pathways were imported into Cytoscape 3.7.0 to obtain a network, which contains 58 nodes and 175 edges, and the nodes increased with the degree value. e results suggested that KHJ may exert an antiinflammatory effect through multicomponent and multitarget. Bergenin, matrine, sophocarpine, calycosin, and trifolirhizin were the main anti-inflammatory active ingredient in KHJ. e main targets of KHJ were EGFR, MMP9, MMP3, MMP1, and PTGS2. e results are shown in Figure 14

Conclusion
Miao medicine was an important part of TCM. However, similar to TCM, the lack of quality standards also seriously restricted the standardization and modernization of Miao medicine. e proposal of Q-Marker pointed out the direction for the quality research of TCM. However, how to discover and identify the Q-Marker was still a great challenge for TCM and Miao medicine.
In this paper, a gray correlation analysis strategy combined with network pharmacology analysis was proposed to investigate the Q-Markers of KHJ.
e results show that bergenin, sophocarpidine, sophocarpine, and trifolirhizin could be regarded as the Q-Markers of KHJ on anti-inflammation. e process of discovering the Q-Markers would provide a promising method of quality control on KHJ. Nevertheless, the specific contribution of each Q-Marker in the formulation had not been clarified, which needs to be further investigated.

Data Availability
All data used to support the findings of this study are included within this paper.

Conflicts of Interest
All the authors declare that there are no conflicts of interest.

Authors' Contributions
Jinpeng Chen, Chengwang Tian, and Tiejun Zhang proposed the concept and design of the study. Jinpeng Chen, Yi Liu, and Xiaohong Gai executed the collecting and analysis of data. Qing Ye and Siyu Zhou accomplished the literature search. Jinpeng Chen drafted the main manuscript, while Chengwang Tian and Tiejun Zhang were responsible for  Evidence-Based Complementary and Alternative Medicine editing and providing critical revision. All authors have read and agreed to the published version of the paper.