Establishment of Quality Evaluation Method for Yinqiao Powder: A Herbal Formula against COVID-19 in China

Yinqiao powder, with significant anti‐inflammatory and antiviral effects, is a classical formula for the treatment of febrile diseases in China. During the SARS period in 2003, Yinqiao powder showed a good antipyretic effect. It also plays a major role in the treatment for COVID-19 in China. Although there are many studies on the chemical compositions and pharmacological effects of Yinqiao powder, there are few studies on the quality standard system of it. In our study, a systematic quality evaluation method of Yinqiao powder combining HPLC fingerprint with quantitative analysis of multi-components by single marker (QAMS) based on network pharmacology and UPLC-Q-Exactive-Orbitrap-MS was established for the first time. In the UPLC-Q-Exactive-Orbitrap-MS experiment, a total of 53 compounds were identified in the extract solution of Yinqiao powder. In addition, 33 blood components were characterized, 23 of which were prototypes. The results of network pharmacology analysis showed that Yinqiao powder may inhibit inflammatory responses by suppressing IL-6, CXCL2, TNFα, NF-κB, etc., in the treatment of COVID-19. The HPLC fingerprint analysis of Yinqiao powder was conducted at 237 nm and 29 characteristic peaks were matched, 11 of which were identified. Forsythoside A was selected as the internal standard reference and double-wavelength (237 nm and 327 nm) was established in QAMS experiment. The repeatability was well under different conditions, and the results measured by QAMS were consisted with that of the external standard method (ESM), indicating that the QAMS method was reliable and accurate. The quality evaluation method of Yinqiao powder would be helpful to evaluate the intrinsic quality of Yinqiao powder more comprehensively, which is conducive to improve the quality standard of Yinqiao powder and provide a beneficial guarantee for the clinical treatment of COVID-19.


Introduction
Yinqiao powder is a well-known traditional Chinese medicine (TCM) formula in China. With signifcant anti-infammatory and antiviral efects, it is clinically used for the treatment of infuenza, infantile pneumonia, hand-foot-mouth disease, etc. [1][2][3]. In particular, during the SARS period in 2003, Yinqiao powder showed good antipyretic efect. Nowadays, coronavirus disease 2019 (COVID- 19) is spreading around the world and causing severe respiratory illnesses and even death.
Yinqiao powder is one of the heat-clearing and detoxifcation prescriptions recommended by the traditional Chinese medicine prevention and treatment plan for COVID-19 in Shaanxi, Jiangsu, Guangdong and Hubei provinces in China [4,5]. However, the quality standard of Yinqiao powder is not perfect at present, so it is necessary to improve the quality control standard of Yinqiao powder.
Yinqiao powder is composed of ten herbal slices with complex ingredients [6]. Te theory of serum pharmacochemistry of TCM believes that only components absorbed into bloodstream are likely to be virtually efective components [7]. Terefore, only by analyzing the serum components after oral administration and determining the direct acting components in the body of Yinqiao powder can fundamentally control the quality of the Yinqiao powder. Te chemical components in TCM and biological samples could be accurately and quickly identifed by UPLC-Q-Exactive-Orbitrap-MS technology, which provides a new method for improving the quality standard of Yinqiao powder [8]. So, UPLC-Q-Exactive-Orbitrap-MS was employed to detect the blood components of Yinqiao powder. In addition, HPLC fngerprint and quantitative analysis of multi-components by single marker (QAMS) are internationally recognized methods. Te comprehensive information of Yinqiao powder could be obtained by HPLC fngerprints [9,10] and QAMS can simultaneously determine the content of various components in Yinqiao powder through one reference substance [11][12][13].
Yinqiao powder is a complex system with multi-components, multi-targets, and multi-action mechanisms through the joint action of various chemical components. Te quality evaluation of it can not only use a single index. In this study, a qualitative and quantitative quality standard evaluation method of Yinqiao powder combining HPLC fngerprint with QAMS was established based on network pharmacology and UPLC-Q-Exactive-Orbitrap-MS technology. Firstly, UPLC-Q-Exactive-Orbitrap-MS was employed to detect the components in Yinqiao powder extract and rat blood. Meanwhile, network pharmacology was used to predict the possible pathways and components of Yinqiao powder in treating for COVID-19. Ten, the common peaks of ten batches of Yinqiao powder were identifed with the existing reference materials and the source of the common peaks was assigned through the establishment of HPLC fngerprint. Finally, the content of the potential active components screened in the frst step was determined by the QAMS method with forsythoside A as the internal standard. Te fowchart of the established analytical strategy is shown in Figure 1.

Chemicals and Reagents.
Cafeic acid, hesperidin, rutin, liquiritin, cynaroside, and forsythoside A were provided by National Institutes for Food and Drug Control (Beijing, China). Arctiin, neochlorogenic acid, chlorogenic acid, isochlorogenic acid A, and isochlorogenic acid C were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). As for phillyrin, it was provided by Chengdu Chroma-Biotechnology Co., Ltd. (Chengdu, Sichuan Province, China). Te detailed information of the above mentioned standard materials is listed in Table S1. All the herbal slices of Yinqiao powder were purchased from Beijing Tongrentang drugstore (Beijing, China), and information of these herbal slices is shown in Table 1.

Prediction of Chemical Components and Related
Targets of Yinqiao Powder. Te chemical components and related targets of Yinqiao powder were predicted using TCMSP (https://tcmspw.com/tcmsp.php) and Swiss Target Prediction platform (https://www.swisstargetprediction.ch). Oral bioavailability (OB) ≥30% and drug-likeness (DL) ≥0.18 were used as the screening criteria. Te active ingredients of Yinqiao powder that are not included in TCMSP or do not meet the screening standards were added according to the Pharmacopoeia of the People's Republic of China (Ch. P) and literature. Te herbal slices-chemical composition-gene data sheet of Yinqiao powder was established based on above information.
Ten, the intersection targets of Yinqiao powder and COVID-19 were obtained using the online platform of Bioinformatics (https://www.bioinformatics.com.cn/) through Venn diagram analysis.

Construction and Analysis of the PPI Network.
Te intersection targets of Yinqiao powder and COVID-19 were submitted to STRING database (https://cn.string-db.org/, Version 11.5) for the construction and analysis of the protein-protein interaction (PPI) network. Te organism was restricted to Homo sapiens and the minimum required interaction score was set as medium confdence (0.400). Ten, the result of TSV format was exported and imported into Cytoscape 3.8.2 software for further analysis.

GO and KEGG Enrichment Analysis.
Te potential genes of Yinqiao powder for the treatment of COVID-19 were imported into Metascape database (https://metascape. org) for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. Species of Homo sapiens and p < 0.01 were set as the flter criteria. Te visual analysis of GO and KEGG enrichment results was performed using Bioinformatics online platform (https:// www.bioinformatics.com.cn/). Te mass spectrometer was operated in both positive and negative ion modes. Te spray voltage was maintained at 3.5 kV in the positive ion mode and 3 kV in the negative ion mode. Full MS spectra were acquired with resolution at 70,000 and AGC target at 1e 6 . MS/MS fragments (spectra) were performed with resolution at 17,500 and AGC target at 1e 5 . Te stepped normalized collision energy (NCE) was set at 20, 40, and 60. Te following parameter settings were used: capillary temperature of 320°C, heater temperature of 300°C, sheath gas velocity of 35 arb, auxiliary gas fow rate of 10 arb, and mass range of m/z 80-1200.
In the fngerprint analysis and QAMS experiments, chromatographic separation was performed on a Termo Syncronis Column (4.6 mm × 250 mm, 5 μm, Termo Fisher, CA, USA). Water with 0.1% phosphoric acid (A) and acetonitrile (B) were used as mobile phase. Te fowing gradient   Serum sample preparation: twelve Wistar rats were divided into an experimental group and a blank group (half of the rats in each group were female). Eight hours before the administration, rats were fasted with free access to water. Rats in the experimental group were given 4 mL of extract of Yinqiao powder, while rats in the blank group were given the same volume of normal saline (0.9% NaCl). 0.5 mL blood samples were collected from the jugular vein of rats at 30 min, 60 min, 90 min, 120 min, and 180 min, respectively after administration. After standing for 1 h, each blood sample was centrifuged at 3,000 rpm for 15 min to obtain the serum. Te serum samples were stored at −80°C until further pre-treatment.
Te freeze-thaw of all serum samples was carried out at 4°C. Each serum sample (200 μL) was mixed with acetonitrile (600 μL), and then centrifuged at 13,000 rpm at 4°C for 15 min. Afterward, the supernatant was dried with a stream of nitrogen at room temperature to obtain the residue. Ten, it was redissolved in 70% methanol-water (100 μL) and centrifuged at 15,000 rpm at 4°C for 15 min. Te supernatant was used for further analysis.

Preparation of Sample Solutions and Negative
Sample Solutions 2.5.1. Sample for Animals. 100 g of Yinqiao powder was prepared according to the Ch. P of the 2020 edition [6]. In a volumetric fask (2000 mL), Yinqiao powder was soaked in 10 times of water for 1 h and fltered after refux extraction for 1 h. Te fltrate was collected, and then the water with residues (1 : 8, w/v) was boiled for an additional 1 h. Two batches of fltrate were mixed. After collecting the volatile oil, the concentration of the fltrate was concentrated to 2 g·mL −1 . Afterward, the fltrate cooled to room temperature was mixed with the volatile oil and the mixture was used in intragastric administration of rats.

2.5.2.
Sample for UPLC-Q-Exactive-Orbitrap-MS. Yinqiao powder (10 g) was extracted according to the above method. Te fltrate was collected and then concentrated to 100 mL. Te mixture of the abovementioned solution (1 mL) and methanol (1 mL) was processed by an ultrasonic assisted extraction (UAE) method for 15 min and then centrifuged at 15,000 rpm at 4°C for 15 min. Te supernatant was isolated for further analysis.

Sample for Fingerprint and QAMS Analysis.
Sample of Yinqiao powder: 0.5 g of Yinqiao powder was accurately weighed and placed in a volumetric fask (25 mL). Te volumetric fask was flled with 70% methanol-water. After being processed by the UAE method for 30 min, the solution of Yinqiao powder was cooled to room temperature and then fltered to obtain the fltrate. Afterward, the fltrate was fltered through a 0.45 μm millipore flter membrane (organic-system) for further analysis.
Preparation of negative control samples: the negative control samples of Lonicerae Japonicae Flos, or Forsythiae Fructus, or Arctii Fructus, or Lonicerae Japonicae Flos and Arctii Fructus were respectively produced as same as "Sample of Yinqiao powder" for further analysis.

Data Processing and Analysis.
Te UPLC-MS data were handled by Compound Discoverer 3.2 software, and compared with ChemSpider, Termo's Chinese Medicine database and mzCloud database. Te HPLC fngerprint analysis was performed using the "Similarity Evaluation System of Traditional Chinese Medicine Chromatographic Fingerprint" software (2004, edition). SPSS17.0 software was used for principal component analysis (PCA) of the key components afecting the quality of Yinqiao powder.

Optimization of Chromatographic Conditions for
Fingerprint and QAMS. We examined diferent mobile phases according to related literatures [14,15]. Te results showed that the baseline was more stable when acetonitrile-0.1% phosphoric acid water was used as mobile phase compared with methanol-0.1% phosphoric acid water.
Terefore, acetonitrile-0.1% phosphoric acid water was chosen as mobile phase. Te elution gradient was based on the conditions in Section 2.3.2. Under this gradient, there were more chromatographic peaks of Yinqiao powder with better resolution and larger peak area.
Furthermore, in order to select detection wavelengths, the chromatograms at 237 nm, 279 nm, 327 nm, and 365 nm were compared. Te results suggested that information collected at 237 nm was the most abundant, and peak area and resolution were better, which could well refect the characteristics of Yinqiao powder in fngerprint analysis. Furthermore, in QAMS experiment, forsythoside A, phillyrin and arctiin had better asymmetry and resolution at 237 nm. Te resolution of neochlorogenic acid, chlorogenic acid, forsythoside A, isochlorogenic acid A, and isochlorogenic acid C were greater than 1.5 without interference of impurity peaks at 327 nm. However, there was no absorption for phillyrin and arctiin at 327 nm. Terefore, the content detection was performed at 237 nm and 327 nm, while 237 nm was selected as the detection wavelength in the fngerprint analysis.

Optimization of Experimental Conditions for Fingerprint Analysis.
Te Yinqiao powder was extracted by the ultrasonic extraction method according to Ch. P [6] and related literatures [16,17]. In our study, 50% methanol, 80% methanol, and 100% methanol were investigated. Te results showed that there was no signifcant diference on the number of components extracted by the three solvents and on the resolution of chromatographic peaks. However, the chromatographic peak area of the samples extracted with 50% methanol was largest under the same chromatographic conditions. Tus, 50% methanol was chosen as the extraction solvent.
In addition, the infuence of solvent peaks on the results must be taken into account [18]. So, 50% methanol was injected according to the above chromatographic conditions. Te solvent peaks of 0-5 min was sheared during peak matching based on the results of the chromatogram. Ten, the chromatograms of 10 batches of Yinqiao powder were matched with common peaks. As a result, there were 112 common peaks under the unrestricted conditions. However, the peak area of most of the common peaks was small and the signal-to-noise ratio (SNR) did not meet the requirements. When the peak area was greater than 0.3, there were 38 common peaks and 11 chromatographic peaks could be identifed. However, some peaks with small area had low resolution and asymmetry. For the further flter, the condition that peak area was greater than 0.4 was carried out. A total of 29 common peaks were obtained, 11 of which could be identifed and the resolution and asymmetry of each common peak were good. In the end, peak area greater than 0.4 was determined as the screening condition.

Optimization of Experimental Conditions for QAMS.
Forsythoside A showed absorption at 237 nm and 327 nm, and the peak shape and resolution were good. Besides, it had high content and stable property in Yinqiao powder.
Terefore, forsythoside A was selected as the internal standard reference.
For the location of chromatographic peaks for each component, relative retention value and retention time diferences are the common qualitative parameters [19]. In our study, the two parameters of each component relative to the internal standard reference were calculated. Finally, neochlorogenic acid and chlorogenic acid were located by retention time diferences while isochlorogenic acid A, isochlorogenic acid C, phillyrin, and arctiin were located by a relative retention value.  [20,21] and spectra.

Analysis of Blood Components in Yinqiao Powder.
A total of 33 blood components were identifed by comparing the chemical components in Yinqiao powder extract, blank serum, and drug serum, 23 components of which were found in both Yinqiao powder extract and drug serum and other 10 components were only present in the drug serum (Table 4). According to related literatures [38][39][40][41], chlorogenic acid, isochlorogenic acid A, isochlorogenic acid C, phillyrin, forsythoside A, arctiin, liquiritin, neochlorogenic acid, cynaroside, rutin, and hesperidin can be absorbed into blood in the form of prototype. It has been reported that   Journal of Analytical Methods in Chemistry 6.27E + 08 [23] phillyrin and forsythoside A have good antiviral and immune regulation efects [42][43][44], neochlorogenic acid, isochlorogenic acid A, and isochlorogenic acid C have antiviral and anti-infammatory activities [45,46], while cynaroside has the efect of inhibiting infuenza virus [47]. Liquiritin, hesperidin, and rutin have obvious anti-infammatory efects [48][49][50]. Te content of the abovementioned components in the Yinqiao powder extract is relatively high, and all of them can be absorbed into the blood with potent anti-infammatory and antiviral efects. Terefore, these components can be used as indicators to further improve the quality standard of Yinqiao powder.

Target Predication of Yinqiao Powder and COVID-19.
A total of 136 components and 294 related targets of Yinqiao powder were obtained according to Ch. P, related literature combined with the TCMSP database. At the same time, GeneCards, OMIM, DisGeNET, and DrugBank databases were used to screen the targets of COVID-19. As a result, 945 targets were fnally retained after removing duplicates. Ten, Venn diagram analysis was performed on the targets of the two and it was found that there were 80 intersection targets ( Figure 2).

Construction and Analysis of the PPI Network.
Te 80 intersecting targets of Yinqiao powder and COVID-19 were submitted to STRING database to construct the PPI network. Ten, the result was imported into Cytoscape 3.8.2 software for further analysis. In the end, a PPI network of Yinqiao powder intreating for COVID-19 with 80 nodes and 2824 edges was acquired, as shown in Figure 3. Te area of the node in the fgure is proportional to the degree value. So, it can be seen intuitively that ALB, IL6, VEGFA, AKT1, TNF, Te chemical components corresponding to the intersection targets were initially identifed as the efective active components of Yinqiao powder. A total of 126 components were found in our study, including quercetin, luteolin, kaempferol, chlorogenic acid, isochlorogenic acid A, isochlorogenic acid C, arctiin, neochlorogenic acid, forsythoside A, liquiritin, and so on. Studies have shown that chlorogenic acid and phillyrin could inhibit the production of TNF-α, IL-1β, IL-6 and alleviate lung infection in mice [43,51]. Neochlorogenic acid could reduce the production of TNF-α, IL-6, and NO, further inhibit the protein expression of iNOS, COX2, TNF-α, IL-6, and attenuate the infammatory response by activating the AMPK/Nrf2 signaling pathway [52]. Arctiin, together with daidzein, glycyrrhizic acid and liquiritin, can inhibit pneumonia by enhancing necroptosis and partial autophagy associated with plc c1 phosphorylation in natural killer cells [53]. Forsythoside A could ameliorate lipopolysaccharide-induced pathological damage, decreased serum levels of TNF-α and IL-6, and inhibited macrophage infltration in the lungs of acute lung injury mice [54].

GO and KEGG Enrichment
Analysis. Furthermore, the 80 targets were submitted to Metascape for GO and KEGG enrichment analysis. P < 0.01 was used as the screening criterion. As a result, 1570 GO items were enriched, including 1413 in biological processes (BP), 55 in cellular components (CC), and 102 in molecular functions (MF). At the same time, 191 KEGG pathways were obtained.
Te GO enrichment analysis results showed that positive regulation of cell migration, regulation of cellular response to stress, response to inorganic substance, infammatory response, and reproductive structure development were involved in BP. In the aspect of CC, it was mainly included membrane raft, endoplasmic reticulum lumen, vesicle lumen, nuclear envelope, transcription regulator complex, etc. Te top 5 items of MF were protein domain specifc binding, protein homodimerization activity, kinase binding, cytokine receptor binding, and RNA polymerase II-specifc DNAbinding transcription factor binding.
Te KEGG enrichment results suggested that the pathways of Yinqiao powder in the treatment for COVID-19 were mainly related to infammation, such as IL-17 signaling pathway, TNF signaling pathway, PI3K-Akt signaling pathway, NF-kappa B signaling pathway, ErbB signaling pathway, coronavirus disease-COVID-19, and so on (Table  S2). COVID-19 is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV 2). SARS-CoV 2 infects alveolar epithelial cells through the angiotensin-converting enzyme 2 (ACE2) receptor, resulting in increased serum levels of free angiotensin II (Ang II). Te increased serum level of free Ang II promotes activation of the NF-kappa B pathway via Ang II type 1 receptor (AT1R), followed by interleukin-6 (IL-6) production. Studies have shown that the serum levels of IL-6, IL-17, and TNF-α in the COVID-19 patients were signifcantly higher than those in the control group [55,56]. Te damage associated with infammatory autoimmune diseases can be reduced by inhibiting IL-17 [57]. Robinson et al. retrospectively explored the potential of anti-TNF in modulating COVID-19-related infammation and concluded that anti-TNF therapies could reduce levels of pro-infammatory cytokines associated with poor COVID-19 outcomes [58]. In addition, NF-κB is involved in the regulation of immunity, infammation and cell survival, which could be activated by TNF-α, IL-1, etc. Numerous studies have demonstrated the potential therapeutic efect of inhibiting the NF-κB pathway in relieving severe forms of COVID-19 [59,60]. Te ErbB signaling pathway can regulate cell proliferation, migration, diferentiation, apoptosis, and cell movement by mediating the PI3K-Akt pathway, JAK-STAT pathway, and MAPK signaling pathway. It was speculated that Yinqiao powder could inhibit JAK-STAT signaling through IL-6, which could reduce infammatory responses and alleviate lung injury [61]. EGFR and EGBB2 are members of the epidermal growth factor receptor family.

Qualitative Analysis by Fingerprint and Multiple Statistical Strategies.
Yinqiao powder and single drug samples were prepared according to the method given in Section 2.5.3, and standard solutions were prepared according to the method given in Section 2.6. Te samples were detected under chromatographic conditions of fngerprint analysis in Section 2.3.2; then, the UPLC fngerprint analysis was performed using the "Similarity Evaluation System of Traditional Chinese Medicine Chromatographic Fingerprint" software (2004, edition). Te matching results are shown in Figure 5.

Precision, Stability, and Repeatability.
Te same batch of Yinqiao powder solution was injected 6 times for the validation of precision and the stability was validated by analyzing the sample of Yinqiao powder at 0, 3, 6, 9, 12, and 24 h at room temperature. It can be seen from the results that RSDs of the retention time of each common peak was 0.01%∼0.08%, and RSDs of relative peak area was 0.32%∼ 2.19%, indicating that the precision of the instrument was excellent and the sample was stable in 24 hours. Meanwhile, six samples of the same batch of Yinqiao powder were accurately weighed, prepared, and tested in parallel to test the repeatability. Te results showed that RSDs of each common peak were less than 0.15% and the relative peak area was less than 2.52%, which indicated that the repeatability was good.

Similarity Evaluation of Fingerprint of Yinqiao Powder.
Ten batches of Yinqiao powder sample were prepared and tested, and the chromatograms of each batch were recorded. Te HPLC fngerprint analysis was performed using "Similarity Evaluation System of Traditional Chinese Medicine Chromatographic Fingerprint" software (2004, edition). S1 was chosen as the reference peak, the fngerprint of Yinqiao powder was established by median method. Te results suggested that the similarity of 10 batches of Yinqiao powder was more than 0.95 (Table S3), indicating that the quality of Yinqiao powder was stable.

Confrmation of Common Peaks in Fingerprint.
29 common peaks were determined and 11 common peaks can be identifed by comparing the chromatograms of 10 batches of Yinqiao powder. Te 11 identifed peaks were as follows: peak 3 was neochlorogenic acid, peak 7 was chlorogenic acid, peak 12 was glycyrrhizin, peak 13 was rutin, peak 14 was  forsythoside A, peak 15 was cynaroside, peak 17 was isochlorogenic acid A, peak 19 was hesperidin, peak 21 was isochlorogenic acid C, peak 25 was phillyrin, and peak 26 was arctiin. Te reference spectrum and fngerprint spectrum of 10 batches of Yinqiao powder are shown in Figures 6(a) and 6(b), the relative retention time is shown in Table S4, and the median retention time and average peak area are shown in Table S5.  Table S6 and Table 5, a total of 9 principal components were obtained when the cumulative rate was 100.000% according to the eigenvalue and variance contribution rate. Te scree plot ( Figure 6(c)) also showed that the slope tends to be gentle after extracting the frst 9 components, indicating that these 9 principal components contain most of the chemical information of Yinqiao powder and they were the key components afecting the quality of Yinqiao powder. Te identifed common peaks were assigned according to the matrix analysis of the frst fve components. Among them, the frst principal components were peak 3 (neochlorogenic acid), peak 7 (chlorogenic acid), peak 12 (liquiritin), peak 17 (isochlorogenic acid A), peak 21 (isochlorogenic acid C), peak 25 (phillyrin), and peak 26 (arctiin). Te second principal components were peak 13 (rutin) and peak 15 (cynaroside). Te third principal components were peak 14 (forsythoside A) and peak 19 (hesperidin). Te abovementioned components were distributed in the frst three principal components, indicating that these components were the key components to ensure the quality of Yinqiao powder, especially the frst principal components.

Partial Least Squares Discriminant Analysis.
Partial least squares discriminant analysis (PLS-DA) was performed by SIMCA 14.1 software with common peak area ( Figure 6(d)). In this model, the VIP value diagram can directly refect the contribution of each chromatographic peak. Te VIP value greater than 1 was usually used as the criterion for screening indicators. Te VIP values of peak 14 (forsythoside A), peak 26, peak 7 (chlorogenic acid), peak 10, peak 2, peak 17 (isochlorogenic acid A), peak 12 (liquiritin), and peak 21 (isochlorogenic acid C) were greater than 1. It was speculated that these components were the key components afecting the quality of Yinqiao powder and played important roles in the quality control.

Quantitative Analysis by QAMS.
In quantitative analysis of Yinqiao powder, the 7 reference standards were selected according to the results of UPLC-Q-Exactive-Orbitrap-MS and network pharmacology. Although the results of the network pharmacology showed that kaempferol, luteolin, (d)   quercetin, etc., may also be the possible components of Yinqiao powder for the treatment of COVID-19, the peak area was relatively small according to the UPLC-Q-Exactive-Orbitrap-MS results. So, it was indicated that the content of these components in Yinqiao powder was low, especially kaempferol, it was not detected in the extract solution of Yinqiao powder.

Investigation of System Applicability and Specifcity.
Te chromatograms of Yinqiao powder samples and negative samples at 237 nm and 327 nm are shown in Figure 7.
3.5.2. Linearity. Two kinds of mixed reference solutions were analyzed, respectively, and the calibration curves were plotted with diferent density solution (X, μg) and corresponding peak area (Y, A). Te results (Table 6) showed that there was a good correlation between the injection volume and peak area for each compound.

Precision, Stability, Repeatability, and Accuracy.
Te precision was measured by continuous injection of two kinds of mixed reference solutions for six times at diferent wavelengths. Te stability was validated by analyzing the sample solution of Yinqiao powder at 0, 3, 6, 9, 12, and 24 h at room temperature. Six identical samples of Yinqiao powder were prepared accurately and peak area at diferent wavelengths were determined in these parallel samples. Te accuracy was determined through adding a certain amount of reference substances to six identical samples of Yinqiao powder with a known content. Te recovery rate was calculated using the equation (1): Recovery rate (%) � Detected amount − Original amount Spiked amount × 100%. (1) As shown in Table S7, the RSD of each component was 0.11%∼2.72% and the recovery rate of seven compounds was 96.32%∼101.79%, indicating that this method could accurately determine the content of seven components in the sample.

Calculation of Relative Correction Factors (RCFs, f x ).
RCFs were calculated using the equation (2): In this equation, W m indicates the concentration of a specifc reference substance, and A m indicates the peak area of reference substance. Forsythoside A was selected as the internal standard reference to calculate the RCFs of phillyrin and arctiin at 237 nm, and the RCFs of    neochlorogenic acid, chlorogenic acid, isochlorogenic acid A and isochlorogenic acid C were calculated at 327 nm (Table 7).

Reproducibility Investigation of RCFs.
Te reproducibility of RCFs was investigated under the condition of diferent instruments and diferent chromatographic columns. Te result showed that the RCFs had a good reproducibility in diferent instruments and columns (Table  S8).

Comparison between QAMS and ESM.
Te relative errors (REs) built by calculating deviations between the QAMS and ESM ranged from −0.83% to 1.08%. Te result showed that there was no signifcant diference on RE of the two methods, indicating that QAMS was feasible to control the quality of Yinqiao powder (Tables 8 and 9).

Conclusions
In this study, network pharmacology and UPLC-Q-Exactive-Orbitrap-MS were combined with HPLC fngerprints and QAMS for the quality evaluation of Yinqiao powder for the frst time. Tis quality evaluation method covered four aspects: chemical compositions, pharmacological efects, qualitative analysis, and quantitative analysis. Network pharmacology and UPLC-Q-Exactive-Orbitrap-MS were used to screen the potential active components of Yinqiao powder. Te network pharmacology results showed that Yinqiao powder may inhibit the infammatory response by suppressing IL-6, CXCL2, TNFα, NF-κB, etc., in the treatment of COVID-19. A total of 53 compounds were identifed in Yinqiao powder extract and 33 blood components were identifed in rat serum using UPLC-Q-Exactive-Orbitrap-MS. Te combination of network pharmacology and serum pharmacochemistry could lay the foundation of the medicinal material for Yinqiao powder. In the HPLC fngerprint analysis of Yinqiao powder, 11 compounds were identifed. It was clear that the main components were from Lonicerae Japonicae Flos, Forsythiae Fructus, and Arctii Fructus by comparing the chromatograms of single herbs. Te multivariate statistical analysis results showed that chlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid C, arctiin, phillyrin, and forsythoside A were key components afecting the quality of Yinqiao powder. For the quantitative analysis of Yinqiao powder, the QAMS method was carried out at double-wavelength (237 nm, 327 nm). Te content of forsythoside A, phillyrin, arctiin, neochlorogenic acid, chlorogenic acid, isochlorogenic acid A, and isochlorogenic acid C was simultaneously determined using forsythoside A as the internal standard reference. Te result of reproducibility investigation of RCFs showed that the reproducibility was good and the RSDs were all less than 3%. In order to verify the reliability and accuracy of the QAMS method established in this experiment, the experimental results measured by QAMS method were compared with those measured by ESM. Te comparison results of QAMS and ESM showed that there was no signifcant diference between the two methods, which indicated that the QAMS method was economical and efcient.
Tis study did not conduct a more in-depth exploration due to the limited time. In conclusion, the quality evaluation method established in our study could comprehensively and scientifcally refect the stability of the internal quality of Yinqiao powder, which is conducive to the improvement of the quality standard of Yinqiao powder and provides a benefcial guarantee for the clinical treatment of COVID-19.
Zhang analyzed data and wrote and revised the manuscript. Huimin Zhang, Jian Song, Qingjun Li, Xinyan Qu, and Ping Wang supervised all work. All authors revised, read, and approved the submitted version.