Predicting and Validating the Mechanism of Qingyi II Granules in the Treatment of Acute Pancreatitis by Network Pharmacology

Network pharmacology, reverse molecular docking, and rat acute pancreatitis (AP) models were used to analyze the mechanism of protection by Qingyi II granules. The chemical components of 7 Chinese herbal medicines in Qingyi II granules were searched through the TCMSP (traditional Chinese medicine systems pharmacology database and analysis platform) database. The active ingredients were screened out in the OB (oral bioavailability) and DL (drug likeness) filters as a condition for inclusion. Then, the prediction analysis of potential targets was performed through databases. A GO (gene ontology) enrichment analysis of target proteins related to AP and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway annotation was performed using the DAVID (The Database for Annotation, Visualization, and Integrated Discovery) database. Finally, the “Herbal-Compound-Target” network was constructed using Cytoscape software. The active component structure and target name were uploaded to the Systems Dock database for reverse molecular docking. With octreotide as a positive control, Qingyi II decoction and Qingyi II granules were administered to AP rats at low, medium, and high doses. The pathological changes in the pancreas were observed using HE staining. The levels of Bcl-2, AMS, BAX, IL-2, and CASP3 in plasma were determined by an ELISA kit. Real-time PCR detected the expression of AKT1 and PIK3CA mRNA in the pancreas. The database predicted 94 active components of Qingyi II granules, 76 potential targets, and 64 signaling pathways. Twenty pathways were directly or indirectly associated with acute pancreatitis, including the TNF signaling pathway and the PI3K-AKT signaling pathway. In the reverse molecular docking experiment, the matching scores of the active components and the target were mainly between 6.0 and 7.0, with strong binding activity. Compared to the normal group, the plasma concentrations of BAX, IL-2, Bcl-2, AMS, and CASP3 in the model group were significantly increased (P < 0.05). Compared with the model group, the low-dose group of Qingyi II granules only significantly reduced IL-2 levels and had no effect on other indicators. The other groups could significantly reduce the levels of AMS, BAX, and CASP3 (P < 0.05). Compared with the model group, the octreotide group and Qingyi II granules high-dose group significantly increased the Bcl-2 level (P < 0.05), and there was no statistical difference in other drug-administered groups. Compared with the normal group, the expression of AKT1 and PIK3CA in the pancreas of the model group was significantly higher. Compared to the model group, the expression of PIK3CA was low in all drug-administered groups. In addition to the low-dose group, the other drug-administered groups significantly reduced the expression of AKT1. Qingyi can reduce the levels of AMS, BAX, IL-2, and CASP3 and increase the levels of Bcl-2. This mechanism may be related to the PI3K- AKT signaling pathway.


Introduction
Acute pancreatitis (AP) is a common clinical condition with rapid onset, rapid progression, and a high mortality rate that can quickly progress to severe acute pancreatitis (SAP). SAP is often accompanied by local or systemic infammatory response syndrome and systemic multiple organ dysfunction [1,2]. Te etiology and pathogenesis of AP are complex and multifactorial pathological processes. Te activation of trypsin and the self-digestion of the pancreas, pancreatic microcirculatory disturbance, infammatory reaction, and intestinal bacterial translocation can cause AP [3,4].
Qingyi II granules are derived from the clinical experience of the Afliated Hospital of Zunyi Medical University. It is composed of Rhei Radix Et Rhizoma (Da Huang), Paeoniae Radix Rubra (Chi Shao), Gardeniae Fructus (Zhi Zi), Aucklandiae Radix (Mu Xiang), Magnoliae Ofcinalis Cortex (Hou Po), Coptidis Rhizoma (Huang Lian), Corydalis Rhizoma (yanhusuo), Moutan Cortex (Mu Dan Pi), and Mirabilite (Mang xiao) [5][6][7]. Da Huang is used as monarch medicine. It is supplemented by Chi Shao, Zhi Zi, Mu Xiang, Hou Po, Huang Lian, yanhusuo, and Mang Xiao. It is combined with various drugs to relieve the depressed liver, clearing away heat, toxic materials, analgesia, and the Tongli attack efect. It is mainly used in spleen and stomach dampness-heat type acute pancreatitis [8]. Many experimental and clinical studies show that Qingyi II granules have a good efect on AP and its complications [9][10][11][12][13]. Network pharmacology is a new discipline based on the theory of systems biology, which analyzes the network of biological systems and selects specifc signal nodes for the design of multitarget drug molecules. Network pharmacology emphasizes multipathway regulation of the signaling pathway to improve the therapeutic efect of drugs and reduce toxicities and side efects, so as to improve the success rate of clinical trials of new drugs and save drug research and development costs [14,15]. Te network pharmacology is a kind of network based on the "drug-target-disease" interaction, using the chemical constituents known in traditional Chinese medicine, combining the existing research results, exploring the multitarget and multiway synergistic efect of the components, and visually displaying the complex action by the network [16]. Reverse molecular docking is a process that uses small molecular compounds (natural products, lead compounds, and chemical compounds) as probes to search the target database of known structures for biological macromolecules that may bind to them. Molecular complexes are formed through spatial and energy matching to predict the potential targets of drugs [17][18][19]. Te holistic and comprehensive characteristics of this research are consistent with the principle of synergy between Chinese medicine and its compound multicomponent, multichannel, and multitarget [20,21]. Terefore, this subject combined with the network pharmacology technology can predict the potential of active components, active targets, and the pathways of Qingyi II granules and further verify the active components and targets by using the reverse molecular docking technique. Te ELISA method and the PCR method are used for the experimental verifcation of the predicted results. Tey provide a reference for the development and clinical application of Qingyi II granules. Te chemical name and number shall be unifed, and the duplicate numerator shall be deleted. By limiting the bioavailability of chemical components of traditional Chinese medicine (OB ≥ 30%) and the drug similarity (DL ≥ 0.18), [22,23] combined with the literature review, the active components in the Qingyi II granule compound were screened.

Te Prediction of Potential Targets of Active.
Te active components were imported into the TCMSP, STITCH, Target-Prediction, and Swiss Target-Prediction databases to obtain their efect targets. After the deduplication was summarized, the target name was input into Uniprot (the limited species was human and existed in rats simultaneously) and converted to a gene name. In the OMIM database, GAD database, TTD database, and PharmGKB database, input keywords "acute pancreatitis" and "Severe acute pancreatitis" to retrieve genes related to acute pancreatitis, remove duplicate and false-positive genes, and match the targets of active components. Te potential target of Qingyi II granule active ingredient in the treatment of AP was obtained.

Construction of the Medicinal Materials-
ActiveComponents-Target Network. Qingyi II granules compound medicinal materials, active components, and target data sets were constructed and introduced into the Cytoscape software. Te medicinal materials, components, and targets were constructed into a ternary comprehensive network model that can refect the medicinal materials, components, and targets using the combined function (Merge) in the software. Te potential targets related to AP were introduced into Cytoscape, a construction componentpotential target network.

Target Pathway Analysis.
Te potential target was imported into the biomolecular function annotation system (DAVID), the "ofcial gene symbol" was selected, the research object was defned as human, and the KEGG pathway annotation and GO enrichment annotation analysis were performed on the AP target. According to the principle of AP correlation reported in the literature, a signifcant enrichment pathway (P < 0.05) related to AP was screened and plotted as a bar graph, and OMIC shares drew a high-level bubble map to screen the key pathways to construct a targetpathway network.

Molecular Docking Verifcation.
Te System Dock (https://systemsdock.unit.oist.jp/) database was used to perform molecular docking of key active components and key targets for the treatment of AP to further validate the target's reliability. Te dock score of the active ingredient and the target protein was obtained by inputting the PDB ID of the target protein and introducing the structure of the active ingredient. Te degree of match between the active ingredient and the target protein was judged based on the dock Score.

Preparation of the Rat AP Model and Grouping
Administration. A total of 56 male Sprague-Dawley rats, weighing 200-250 g, were fed for one week. Tey were randomly divided into seven groups, eight in each group, including the control group, model group, octreotide group, Qingyi II decoction group, and Qingyi II granules in the low, medium, and high-dose groups. Tey were fasting for 12 h, drinking only water. AP rats were induced with intraperitoneal injections of caerulein (CAE) 60 μg/kg every 1 h for 6 h, the fnal intraperitoneal injection of CAE, and intraperitoneal injections of lipopolysaccharide (LPS) 15 mg/kg. After successful modeling, octreotide group (100 μg/kg), Qingyi II decoction group (13 g/kg), and Qingyi II granules in low-dose groups (6.5 g/kg), medium-dose groups (13 g/kg), and high (26 g/kg) dose groups, the three groups were dosed once a day for fve consecutive days. Octreotide was injected subcutaneously, and the other groups were intragastrically administered. Te control and model groups were given the same amount of normal saline.
Blood was collected from the abdominal aorta after the last administration, and the plasma was taken by centrifugation (5000 rpm, 10 min) and stored in a refrigerator at −20°C. After the rats died, the pancreas was quickly removed, placed in a Petri dish containing physiological saline, rinsed, and blotted dry using flter paper. Te same portion of the pancreas was cut into a 10% formaldehyde solution for fxation. Te gradient alcohol was dehydrated, made transparent with xylene, serially sliced with parafn, and subjected to HE staining to observe pancreatic tissue lesions. Te ELISA kit detected the levels of Bcl-2, AMS, BAX, IL-2, and CASP3 in plasma. Real-time PCR detected the expression of AKT1 and PIK3CA mRNA in the pancreas.
Statistical analysis was performed using SPSS 17.0 software. Te diferences were expressed as the mean-± standard deviation. A one-way analysis of variance was used between groups. Te LSD test was used to compare groups with variance, and Dunnett's T3 test was used for groups with irregular variance. Dunnett's T3 test was performed for those with irregular variance. P < 0.05 indicated that the diference was statistically signifcant.

Screening of Active Components and Targets of Qingyi II
Granule in the Treatment of AP Te selected Qingyi II granules, active ingredients, and target were sequentially introduced into Cytoscape software to construct a Qingyi II granules-activeingredient-target network ( Figure 1). Each edge in the fgure represents the interaction between the drug and the compound or the compound and the target. Te greater the efect in a network, the more edges connected to it, and the higher the degree of the node.

Qingyi II Granules for the Treatment of AP Component-Target Network.
A total of 1023 known AP-related genes were found by the OMIM, GAD, TTD, and PharmGKB databases, and a total of 1012 were obtained after deleting the duplicate genes. When 76 targets from Figure 1 were compared, 94 active components in Qingyi II granules were applied to these targets. Possible therapeutic targets in this formulation are shown. Te active compounds and potential targets were constructed to construct an anti-APcompound-target interaction network ( Figure 2). It contains 170 nodes (94 compounds and 76 target sites) and 561 edges. In Figure 2, MOL000098 (Quercetin) has the highest degree (34 targets), which makes it the central node of the network, derived from Cortex Moutan, Gardenia, and Rhizoma Corydalis. Tis suggests that quercetin may be a key active ingredient in the efcacy of Qingyi II granules, followed by MOL000422 (kaempferol), MOL000790 (isocorypalmine), MOL000217 ((S)-Scoulerine), and MOL000472 (emodin), which had 18, 16, 16, and 15 targets, respectively. Quercetin and Kaempferol belong to favonoids, isocordierquat and (S)scoulerine belong to alkaloids, and emodin belongs to anthraquinone, which can inhibit the release of infammatory factors, sterilization, and antivirus, etc [24][25][26]. It is indicated that favonoids, alkaloids, and anthraquinones may be essential components of the clear Qingyi II granules. In the network, prostaglandin G/H synthase 2 (PTGS2) and androgen receptor (AR) have 39 and 38 compounds, respectively. Tere were 29 thrombin and acetylcholinesterase (ACHE), 27 mitogenactivated protein kinase 14 (MAPK14), peroxisome proliferator-activated receptors (PPARs), and trypsin 1 (PRSS1). Twenty-six compounds work together, and it is speculated that it may be a key target for AP treatment with Qingyi II granules, refecting the characteristics of multiple components and multi-target interactions of Qingyi II granules.

Target Biological Process Analysis.
Using the DAVID database for GO enrichment analysis (Figure 3), the results are shown in Figure 3. Te results show that biological processes are closely related to the positive regulation of cell proliferation, the apoptosis process, signal transduction, and the infammatory response, refecting that AP pathogenesis is related to multiple biological processes in vivo. It also shows that the Qingyi compound may play a role in treating AP by improving these biological processes.

Target Signal Path Analysis.
Te potential targets were imported into the DAVID database for KEGG analysis, and 20 signaling pathways with a high correlation with AP were selected (P < 0.05), as shown in Table 1. Te high-level bubble map is drawn by OMIC share, as shown in Figure 4. Te enrichment degree of the diferentially expressed genes in the pathway is refected. Te enrichment degree of KEGG is compared by the number of genes on the pathway, the enrichment factor (Rich factor represents the ratio of the number of genes located in this pathway entry to the total number of all genes located in this pathway entry), and P value. Te greater the rich factor, the greater the degree of enrichment.
Six key pathways for the treatment of AP-potential targets were screened out with AP-related pathways (P < 0.05). Te Cytoscape software was used to construct a target-path network, as shown in Figure 5. Nine potential targets are acting on the TNF and PI3K-AKT signaling pathways; six targets synergistically act on the T-cell receptor signaling pathway and the toll-like receptor signaling pathway; participating in the p53 signaling pathway; and fve targets of the NF-κB signaling pathway. Tis indicates that the Qingyi II granules may achieve their purpose of treating AP by regulating these signaling path-related genes.

Analysis of Molecular Docking Verifcation Results.
Te target of the previously predicted pathway is used for molecular docking with the primary active ingredient in the formulation. Aloe Aloe-Emodin, Rhein, Emodin, Physcion, Paeonol, Geniposide, and Costunolide were ligated with IL-2, Bcl-2, BAX, CASP3, AKT1, and PIK3CA, respectively. Te degree of matching between the component and the target protein is determined by the docking score value. [27] Te results are shown in Table 2. Te score value above 4.25 indicates that there is a certain amount of binding between the active ingredient and the key target. If the value is greater than 5.0, it indicates better activity, and greater than 7.0 indicates strong binding activity. [28] From the table below, the docking score between the active ingredient and the target is higher than 4.25, mainly concentrated between 6.0 and 7.0. Te results indicate that the active ingredient has a good binding activity to the target protein. Tis result verifes the reliability of the predicted target. Tese targets can be used to detect pharmacodynamic indicators and for further experimental verifcation. Figure 6, edema, slight degeneration and necrosis of acinar epithelial cells, and infltration of interstitial cells occurred in the model group when compared to the blank group, indicating that the disease in the model group changed signifcantly. Compared to the model group, the pathological changes in the low-dose group were not signifcantly improved. Te positive control group, the decoction group, the middledose group, and the high-dose group were improved to varying degrees, suggesting the poor therapeutic efect of the low-dose group and the decoction group. Te middle-dose group and the high-dose group have a certain therapeutic efect. 4 Evidence-Based Complementary and Alternative Medicine  Evidence-Based Complementary and Alternative Medicine 5

Efect of AMS, IL-2, Bcl-2, BAX, and CASP3 on Plasma in
Rats. As shown in Figure 7, the plasma BAX, IL-2, Bcl-2, AMS, and CASP3 concentrations in the model group, were signifcantly higher than those in the blank control group (P < 0.05). Compared to the model group, the low-dose group of Qingyi II granules only signifcantly reduced IL-2 levels and did not afect other indicators. Compared with the model group, except for the low-dose group of Qingyi II granules, all the other administration groups could significantly decrease the levels of AMS, CASP3, and BAX (P < 0.05). Compared with the model group, the octreotide group and the high-dose group of Qingyi II granules signifcantly increased the level of Bcl-2 (P < 0.05), but there was no signifcant diference in other administration groups.  Compared with the normal group (Figure 8(a)), the expression of AKT1 mRNA in the model group was signifcantly increased (P < 0.05). Tere was no statistically signifcant diference between the low-dose group and the model group, but the expression of AKT1 mRNA in the other administration groups was signifcantly lower than that in the model group (P < 0.05). Compared to the normal group (Figure 8(b)), the expression of PIK3CA mRNA in the model group was signifcantly increased (P < 0.05). Compared to the model group, PIK3CA mRNA was not statistically signifcant in all the administration groups.

Conclusion and Discussion
Network pharmacology utilizes the analysis of pathway targets related to drugs and diseases to predict the therapeutic efect of drugs, which has the potential to improve clinical trials of new drugs. Te success rate and the advantage of saving drug research and development costs [29]. By referring to the network pharmacology evaluation method guide, the data information of diferent platforms was collected and compared. Analyze and use this method to predict the mechanism of Qingyi II granules in the treatment of acute pancreatitis and guide subsequent experimental verifcation [30,31].
In clinical practice, Qingyi II decoction is used as medicine. Although decoction is suitable for the principle and fexibility of TCM syndrome diferentiation and treatment, plus or minus symptoms, it is bulky, easy to deteriorate, and inconvenient to take, carry, and store. Te previous research of our research group shows that after Qingyi II decoction is changed into Qingyi II granules, it can not only ensure the curative efect but also be convenient for patients to take, carry, and store and is superior to the decoction in terms of quality control and preparation stability. Te etiology and pathogenesis of acute pancreatitis are complex multifactor pathological processes. Qingyi II granules are efective in the treatment of spleenstomachdampness-heat acute pancreatitis. However, its mechanism is not clear. Te integrity and comprehensiveness of the network pharmacology research strategy are consistent with the principle of multicomponent, multichannel, and multitarget synergy of traditional Chinese medicine and its compounds [32]. It is benefcial to explain the mechanism of AP treatment systematically and comprehensively.
Tis study discovered that Aloe-Emodin, Rhein, Emodin, and Physcion in Qingyi II granule compounds could directly act on PIK3CA, MAPK14, or IL-2, IL-1B, IL-6, and TNF via component and target prediction. Te targets of PAEONOL are BCL-2, IL-2, and TNF. Geniposide can act on BCL-2, Costunolide, BAX, and CASP3. Tese targets are important for predicting six signaling pathways. Tese six signaling pathways have been reported in the literature. A PI3K-AKT signaling pathway is downstream of the toll-like receptor pathway and the TNF signaling pathway, but upstream of the NF-κB and the p53 signaling pathways in the pathogenesis of AP.
Te PI3K/AKT pathway is an important information pathway in the pathogenesis of AP. Current studies have confrmed that PI3K/AKT can treat acute pancreatitis through multitarget, multipathway, and multifunction. Tere are three types of PI3K: I, II, and III. Among them, the most widely studied class I PI3Ks are composed of the regulatory subunit P85 and the catalytic subunit P110. PI3K is the intermediate bridge of a variety of cellular responses and plays a key role in cell apoptosis, pyroptosis, autophagy, oxidative stress, and other pathways under the infuence of a variety of upstream or downstream factors. AKT, also known as protein kinase B (PKB), is a serine/threonine protein kinase composed of AKT1, AKT2, and AKT3 subtypes. Activated PI3K phosphorylates AKT at phosphorylation sites Ser473 and Tr308, and AKT activates or inhibits its downstream targets such as NF-κB and mTOR to participate in the growth and survival of the body [33]. Te PI3K/Akt signaling pathway is involved in the physiology and pathology of various tumors and can signifcantly inhibit the proliferation, migration and metastasis of cancer cells [34,35]. It has been reported that the PI3K-AKT signaling pathway can efectively regulate the release of infammatory factors to achieve AP prevention and treatment [36]. Terefore, the PI3K-AKT signaling pathway plays an important role in the treatment of AP by Qingyi II granules. Te results of the molecular docking experiments show that there is good binding activity between the main active ingredients and the key target [37]. Te reliability of the predicted target was further verifed. So, we selected this pathway and the downstream targets AMS, IL-2, Bcl-2, BAX, and CASP3 for pharmacodynamic validation. Te modeling method used in this experiment is simple, and its success rate is high. Tere are many methods to establish the acute pancreatitis model. CAE can only induce mild acute pancreatitis. In this experiment, CAE combined with LPS was selected to establish the SAP model. After modeling, it was observed that the surface of the pancreas was dark red, with hyperemia and edema. Under the microscope, the pancreatic tissue had cell degeneration and necrosis, edema, and interstitial infammatory cell infltration. Te serum amylase was signifcantly increased by abdominal aortic blood collection. Te model was successfully established.
Te results of the pharmacokinetics test showed that Qingyi II granules could up-regulate the expression of AKT1 mRNA, decrease the concentration of AMS, IL-2, BAX, and CASP3, and increase the concentration of Bcl-2. PI3K-AKT is an important mediator for regulating the expression of NF-κB and downstream genes. Several studies have confrmed its vital role in the occurrence and development of acute pancreatitis. It is speculated that the mechanism of Qingyi II granules in the treatment of acute pancreatitis may be through inhibiting AKT1 phosphorylation, decreasing Bax, increasing Bcl-2 expression inhibiting the activation, and release of CASP3, and inhibiting apoptosis.
In conclusion, it is correct to study the mechanism of the Qingyi II granule in the prevention and treatment of acute pancreatitis using the combination of network pharmacology and molecular docking technology. Tis reveals the characteristics of multicomponent, multitarget, and multichannel treatment of Qingyi II granules. It also embodies the principles of integrity and systematicity in the treatment of diseases with traditional Chinese medicine. It is of great signifcance to explore the targeted regulation of the PI3K/ AKT signaling pathway in the prevention and treatment of AP from the perspective of related targets and cell functions of the PI3K/AKT signaling pathway and traditional Chinese medicine treatment of AP, which is expected to provide new ideas for basic research and clinical treatment of AP. Te synergistic efects of the multiple compounds in Qingyi II granules and the relationship between the efects of these compounds and their mechanisms in AP will be investigated in the future.

Data Availability
Te data included in this study is available upon reasonable request to the corresponding author.

Conflicts of Interest
Te authors declare that there are no conficts of interest.