Anti-Inflammatory Effects and Mechanisms of Pudilan Antiphlogistic Oral Liquid.

Pudilan antiphlogistic oral liquid (PDL) is a commercial traditional Chinese medicine widely used in the treatment of a variety of inflammatory diseases. However, the specific mechanisms of PDL's anti-inflammatory effects have not been fully understood. In this research, five classic inflammatory models and a network pharmacology-based strategy were utilized to evaluate its anti-inflammatory efficacy and elucidate its multicomponent and multitarget mode of the anti-inflammatory mechanism. A systems pharmacology approach was carried out via a holistic process of active compound screening, target acquisition, network construction, and further analysis. The potential component-target-associated anti-inflammatory mechanisms of PDL were further verified both in vivo and in vitro. The results showed that PDL exhibited a proven anti-inflammatory effect on multiple types of inflammatory models, including β-hemolytic streptococcus-induced acute pharyngitis, LPS-induced acute lung injury, xylene-induced ear swelling, carrageenan-induced paw edema, and acetic acid-induced capillary permeability-increasing models. Systems pharmacology analysis predicted 45 ingredients of PDL that interact with 185 targets, of which 38 overlapped with the inflammation-related targets. Furthermore, KEGG pathway analysis showed that the predicted targets were mainly involved in hypoxia-inducible factor (HIF)-1, tumor necrosis factor (TNF), nuclear factor kappa-B (NF-κB), and NOD-like receptor (NLR) pathways. Both in vivo and in vitro experiments clarified that PDL has anti-inflammatory potency by inhibiting PI3K and p38 phosphorylation and activating the NLRP3 inflammasome. Our results suggested that PDL has an efficient and extensive anti-inflammatory effect, and its anti-inflammatory mechanisms may involve multiple inflammatory-associated signaling pathways, including HIF-1- and TNF-mediated pathways and NLRP3 inflammasome activation.

Introduction

Inflammation is clinically manifested as a reddened, swollen, hot, and usually painful condition.[1] In order to eliminate stimuli and promote tissue repair and healing, innate and adaptive immunological cells are activated and recruited, contributing to local or systemic inflammation in the body.[2] It is a fundamental protective response of the body. Still, excessive inflammation would cause damage to the body, leading to disorders like type 2 diabetes, asthma, Alzheimer’s disease, rheumatoid arthritis, colitis, cancer, and so forth.[3−7] The most commonly used anti-inflammatory drugs are chemical synthetics, including steroidal anti-inflammatory drugs (SAIDs) and non-SAIDs. However, these potent therapeutics are often accompanied by irritating adverse effects such as obesity, hormonal disorders, aggravated infections, gastrointestinal irritation, liver damage, and so forth.[8,9] Because anti-inflammatory drugs are a significant class in clinical treatment, second only to anti-infectious agents, developing anti-inflammatory drugs with potent efficacy and negligible side effects is an urgent issue for medical workers to deal with.

Complementary and alternative medicine, especially traditional Chinese medicine (TCM), is widely used for anti-inflammatory purposes. Inflammation is recognized as the “fire evil” in the theory of TCM, which requires herbs of cold and calm nature to tackle. Pudilan antiphlogistic oral liquid (PDL), collected in Chinese Pharmacopoeia, was a well-known TCM formula prepared from four herbs of cold and calm nature, including Scutellaria baicalensis Georgi, Isatis tinctoria L., Corydalis bungeana Turcz, and Taraxacum mongolicum Hand.-Mazz. PDL has been clinically used for inflammatory diseases such as pharyngitis, tonsillitis, and other upper respiratory infection treatments.[10] Previous reports showed that PDL had a therapeutic effect on pneumonia in mice infected with Streptococcus pneumoniae(11) and had the effect of enhancing immunity of immunocompromised mice.[12]

Nevertheless, investigation on the underlying mechanisms of the potential protective effects of PDL against inflammation has been quite limited. Considering the complex composition and multitarget interactions of TCM, it is extremely difficult to conduct a systematic and in-depth study on the effects and mode of action of PDL. Systemic pharmacology is built to study the interaction between drugs and the body from a systemic level, which helps to study the discipline and mechanisms of the action of the drugs, especially some ancient but potent formulae.[13] Studies have found that the construction of complex biochemical networks of living organisms such as cells presented a large-scale period from transients (such as metabolic reactions) to long-term regulation (such as cell development), from different localities within cells to tissues and organs.[14] Large-scale leaps will also present multiple dimensions of the three-dimensional network from small chemical molecules, genes to proteins.[15] Therefore, systemic pharmacology can solve a series of problems brought about by the complexity of TCM, which provides a new approach to predict the active ingredients and candidate targets through a holistic process of active compound screening, target “fishing”, component–target–disease (C–T–D) network construction, and target enrichment analysis.[16−19]

Therefore, in this study, an innovative and systematic approach was used to predict the anti-inflammation targets of PDL comprehensively. We investigated the effect of PDL on several inflammation models and attempted to explore the underlying mechanism of PDL on inflammation via systems pharmacology analysis, and then subsequent experimental validation was presented in vivo and in vitro.

Results and Discussion

PDL Markedly Alleviated the β-Hemolytic Streptococcus-Induced Inflammatory Response in Acute Pharyngitis Rats

An acute pharyngitis rat model was established as shown in Figure A. As shown in Figure B, the expression of IL-6 and IL-1β in serum significantly increased after β-hemolytic streptococcus infection, whereas PDL administration could significantly downregulate these pro-inflammatory factors. Furthermore, pathological staining showed that the pharyngeal mucosa of control rats was covered with multiple layers of squamous epithelium, with thin layers of connective tissue under the epithelium and no degeneration or necrosis of the epithelial cells after the pharyngeal infection with β-hemolytic streptococcus. In contrast, the hypopharyngeal wall tissue of the pharyngeal mucosa of model rats was infiltrated by mononuclear macrophages and neutrophils, accompanied with local hyperemia and edema. More importantly, PDL administration could relieve the infiltration of inflammatory cells in the pharyngeal tissue (Figure C,D). Collectively, PDL dramatically alleviated inflammatory responses in the acute pharyngitis rat model, in consistence with the clinical efficacy of PDL.

Figure 1

Therapeutic effects of PDL on rats with acute pharyngitis. (A) Flow chart of β-hemolytic streptococcus-induced acute pharyngitis model in rats and dosage regimen. Pharyngeal mucosa of rats was injected with 4 × 108 CFU β-hemolytic streptococcus in 0.1 mL of PBS on day 5 and 6. Amoxicillin at 0.18 g/kg and PDL at different dosages (0.9, 2.7, and 8.1 g/kg) were administered for 7 days in the corresponding groups. Furthermore, 24 h after the last infection, serum and pharyngeal mucosa samples were collected. (B) Pro-inflammatory cytokines IL-6 and IL-1β in serum were detected using ELISA (##P < 0.01 vs the control group and *P < 0.05 and **P < 0.01 vs the model group). (C,D) pharyngeal mucosa sections (5 μm) were processed according to a standard HE staining protocol, and histopathological changes were observed (n = 10, magnification: ×100, scale bar = 100 μm).

Effect of PDL on LPS-Induced Acute Lung Injury (ALI) in Mice

To confirm the effects of PDL (1.3, 3.9, and 11.7 g/kg) on inflammation of the lower respiratory tract, we examined the effects of PDL on LPS-induced ALI in mice (Figure A). The results showed that PDL (11.7 g/kg) reduced the wet/dry ratio of mouse lung tissue (Figure B). Meanwhile, PDL significantly restrained the activity of myeloperoxidase (MPO) in mice (Figure C) and reduced the level of IL-6, IL-1β (P < 0.01, P < 0.01, and P < 0.01), and TNF-α in bronchoalveolar lavage fluid (BALF) (Figure D–F). PDL could significantly reduce the number of neutrophils in BALF (Figure G,H). After hematoxylin and eosin (HE) staining of the lung tissue and then pathological scoring was performed, the result showed that 11.7 g/kg PDL significantly alleviated the inflammatory infiltration of the lung tissue and reduced blood exudation (Figure I,J).

Figure 2

Effect of PDL on LPS-induced ALI in mice. (A) Experimental design. (B) Wet/dry ratio of mice lung tissue. (C) MPO activity of mice lung tissue. (D) IL-6 in BALF. (E) IL-1β in BALF. (F) TNF-α in BALF. (G) Total number of cells in BALF. (H) Neutrophils count in BALF. (I) HE staining of the lung tissue, red arrowheads indicate the thickened bronchial wall (n = 9–12, scale bar = 100 μm; magnification: ×200). (J) ALI score. Data are the mean ± SD, n = 8–12, ##P < 0.01 vs control, *P < 0.05 vs model, and **P < 0.01 vs model.

Effect of PDL on Xylene-Induced Ear Swelling in Mice

Mice were given PDL or aspirin (ASA, 0.2 g/kg) for 7 days by intragastric administration (Figure A). Except for the control group, the other groups were stimulated with 30 μL of xylene to establish the acute ear swelling model (Figure A). The ear thickness was measured using a thickness gauge, and the ear tissue was microscopically examined after HE staining. It showed that PDL significantly decreased the ear swelling of mice (Figure B). The results of the lymphocyte count showed that PDL significantly reduced the number of inflammatory cells infiltrating the ear tissue (Figure C). HE staining showed that 3.9 and 11.7 g/kg PDL could substantially improve the inflammatory infiltration and ear swelling (Figure D).

Figure 3

Effects of PDL on xylene-induced ear swelling in mice. (A) Experimental design. (B) Ear swelling of mice. (C) HE staining of mice ear tissues, red arrowheads indicate the thickened ear (n = 10, scale bar = 100 μm; magnification: ×100). (D) Statistics of inflammatory cells. Data are the mean ± SD, n = 10, ##P < 0.01 vs control, *P < 0.05 vs model, and **P < 0.01 vs model.

Effect of PDL on Carrageenan-Induced Paw Edema in Rats

Rats were given PDL or ASA (0.2 g/kg) for 7 days by intragastric administration. Except for the control group, the rats replicated the acute paw edema model by the subcutaneous injection of carrageenan (Figure A). The paw volume of the rats was measured every hour, and the change curve of edema rate was drawn in Figure B. The results showed that 2.7 and 8.1 g/kg PDL significantly inhibited the paw edema rate of rats 2 h after administration. During the period of 3–7 h, each dose of PDL significantly inhibited the paw edema rate of rats (Figure C).

Figure 4

Effects of PDL on paw edema of rats induced by carrageenan. (A) Experimental design. (B) Statistics of paw edema changes. (C) Paw edema in different time points. Data are the mean ± SD, n = 10–11, ##P < 0.01 vs control, *P < 0.05 vs model, and **P < 0.01 vs model.

Effect of PDL on Acetic Acid-Induced Evans Blue Leakage

Mice were given PDL and ASA (0.2 g/kg) for 7 days by intragastric administration. All mice were injected with 2% Evans blue physiological saline solution intravenously. Mice were subsequently injected intraperitoneally with 0.6% acetic acid saline solution to replicate the capillary permeability-increasing model (Figure A), and mice in the control group were injected with saline. The results showed that PDL significantly inhibited the increase in the capillary permeability of mice induced by acetic acid (Figure B).

Figure 5

Effects of PDL on acetic acid-induced Evans blue leakage. (A) Experimental design. (B) Evans blue leakage from mouse peritoneal lavage fluid. Data are the mean ± SD, n = 10, ##P < 0.01 vs control, and **P < 0.01 vs model.

Predicted Active Components and Targets of PDL against Inflammation

To elucidate the mechanism of PDL against inflammation, we first utilized a systems pharmacology approach to predict the active components in PDL and their potential targets. As a result, 370 chemical ingredients of the four herbal medicines in PDL were retrieved from the traditional Chinese medicine systems pharmacology (TCMSP) database and related literatures, including 143 in S. baicalensis Georgi, 169 in I. tinctoria L., 29 in C. bungeana Turcz, and 29 in T. mongolicum Hand.-Mazz. As listed in Table S1, 45 ingredients were screened out as the active virtual candidates in PDL following the criteria OB ≥ 30%, DL ≥ 0.18, A log P ≥ 5, and Caco-2 ≥ 0.4. Most of the 45 ingredients were validated in previous studies.[20−22]

According to the target prediction system and several databases, we retrieved 185 potential targets out of the predicted 45 ingredients in PDL. Besides, 471 targets were screened out based on the keyword “inflammation” (Figure S1), and there were a total of 38 intersections between PDL targets and inflammation-related targets (Figure S2).

Network analysis provides an efficient tool for strengthening our understanding of the action mechanisms of multicomponent and multitarget TCM regulatory networks. As a result, an herb–component–target–disease (H–C–T–D) network of PDL was composed of 88 nodes and 290 edges. The top 10 ingredients of the degree value were quercetin, wogonin, luteolin, baicalein, corynoline, indirubin, acacetin, moslosooflavone, coptisine, and coryincine. Also, the top 10 targets were PTGS2, NOS2, PIK3CG, F2, NOS3, PPARG, F7, ATP7B, C5, and BCL2 (Figure A).

Figure 6

Network pharmacology analysis of PDL against inflammation (A) H–C–T–D network interaction of PDL. The H–C–T–D network of PDL was established using Cytoscape 3.7.1, which helped us identify the corresponding targets of each ingredient in PDL and get an overview on the potential modes of action of PDL on inflammation. (Red triangle: herb; purple diamond: component; orange square: target; and green circle: disease). (B) KEGG pathway analysis of the PDL-targeting genes. The DAVID Bionformatics Resources 6.8 System was utilized to perform the enrichment analysis of the potential target genes of PDL on inflammation, and the pathways with p-values ≤ 0.05 were considered for better prediction and were selected for further verification of the mechanisms of PDL against inflammation.

Afterward, KEGG pathway analysis was carried out on the predictive target genes of PDL, through which we selected the pathways according to the P-value for further study. As shown in Figure B, the top 18 pathways were listed. PDL might regulate inflammation mainly through hypoxia inducible factor-1 (HIF-1), tumor necrosis factor (TNF), nuclear factor kappa-B (NF-κB), and NOD-like receptor (NLR) pathways, and so forth.

According to the DAVID system, PI3K and p38 play pivotal roles in HIF-1 and TNF signaling, the top two pathways in our enrichment analysis (Figure S3).

Effect of PDL on Inflammation was through PI3K

We next verified the efficacy of PDL regulating these two targets on a protein level in the β-hemolytic streptococcus-induced acute pharyngitis model. As shown in Figure , PDL had negligible effects on total PI3K and P38, whereas their corresponding phosphorylated protein were both downregulated by PDL. It indicated that PDL might exert its anti-inflammatory influence by restraining the activation of PI3K and its downstream protein P38, in accordance with the prediction by systems pharmacology.

Figure 7

Efficacy of PDL on PI3K and p38 in the acute pharyngitis rat model. Pharyngeal mucosa tissues from different experimental groups were collected 24 h after the last infection of β-hemolytic streptococcus. PI3K, p38, and their corresponding phosphorylated proteins were determined using Western Blot (WB). The data are representative of three independent experiments.

Effects of Wogonin and Corynoline on the NLRP3 Inflammasome Priming Phase

An NLR pathway is also one of the potential pathways for PDL to exert anti-inflammatory effects. The role of the NLRP3 inflammasome in various diseases has become the focus of attention recently. The NLRP3 inflammasome can identify pathogen-associated molecular patterns or danger-associated molecular patterns and, in turn, mediates host cells’ immune response to pathogens and cell damage.[23] Activation of NLRP3 inflammasomes requires two steps. Transcriptional levels of NLRP3, pro-IL-1β, and pro-caspase-1 can be elevated by infection, stress, or injury inducers as the first signal through the NF-κB pathway.[23] ATP, protozoa parasite, crystal or other substances (uric acid salt, calcium oxalate, silica, asbestos, etc.), and metabolites in the body can be used as the second signal,[23] makes NLRP3 protein oligomers, ASC, and caspase 1 form the NLRP3 inflammasome. The NLRP3 inflammasome can cut pro-IL-1β into mature IL-1β, which released into extracellular would induce inflammation.[24] The NLRP3 inflammasome plays a key role in inflammation and has gradually become one of the therapeutic targets. Systems pharmacology predicted that quercetin, wogonin, luteolin, baicalein, corynoline, indirubin, and acacetin in PDL were important potential active components. According to the composition analysis, quercetin, wogonin, luteolin, baicalein, and corynoline were the active components with the high content in PDL (Table S1). The effects of baicalein, quercetin, and luteolin on the activation of the NLRP3 inflammasome have been reported,[25,26] but wogonin and corynoline have been rarely reported. We wondered whether these two components could inhibit the activation of the NLRP3 inflammasome to achieve the inhibitory effect of IL-1β.

3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) results showed that wogonin had no significant effect on the cell activity of primary peritoneal macrophages at 0.1–100 μM, and corynoline had no significant impact on the cell activity of peritoneal macrophages at 0.1–10 μM, while inhibited cell proliferation at 100 μM (P < 0.01) (Figure A,B). Before stimulating peritoneal macrophages with LPS and ATP, we pretreated wogonin (10 μM) and corynoline (1 μM), which dissolved in DMSO, for 1 to 24 h. The level of IL-1β was drastically reduced by wogonin (1, 10, and 100 μM) and corynoline (0.1, 1, and 10 μM), respectively. Then, we tested the onset time of wogonin and corynoline. The level of IL-1β was significantly inhibited by 10 μM wogonin and 1 μM corynoline after incubation for 1–24 h (Figure C,D).

Figure 8

Effects of wogonin and corynoline on the NLRP3 inflammasome priming phase. (A) Effect of wogonin on the viability of peritoneal macrophages. (B)Effect of corynoline on the viability of peritoneal macrophages. (C) Effect of pretreated wogonin (0.01, 0.1, 1, 10, and 100 μM) and corynoline (0.001, 0.01, 0.1, 1, and 10 μM) on the IL-1β release level. (D) IL-1β production of peritoneal macrophages pretreated wogonin (10 μM) and corynoline (1 μM) for 1, 4, 12, and 24 h, separately. (E,F) After stimulating with LPS for 4 h and ATP for 0.5 h, peritoneal macrophages were treated with wogonin (0.1, 1, 10, and 100 μM) or corynoline (0.01, 0.1, 1, and 10 μM). (E) IL-1β level in cell supernatant. (F) Effects of wogonin and corynoline on the activation of NF-κB and the downstream protein expression level of NLRP3 and Pro-IL-1β. Data are the mean ± SD, n = 3, ##P < 0.01 vs control, and **P < 0.01 vs LPS + ATP + DMSO.

There are two signaling stages in the activation of the NLRP3 inflammasome, which are stimulated by LPS and ATP.[27] The first signal is initiated by different TLR ligands, including the expression of NF-κB transcription and downstream NLRP3 and pro-IL-1β protein, induced by the LPS danger signal.[28] The second signal is stimulated by ATP, which contributes to the formation of NLRP3 inflammatory complex, the activated caspase-1 shears pro-IL-1β into activated IL-1β, which leads to inflammation in the body.[29] To investigate the action stage of wogonin and corynoline, we administrated wogonin and corynoline after the stimulation of LPS, then stimulated by ATP. The results showed no significant change in the release level of IL-1β (Figure E), indicating that the inhibition of IL-1β release by wogonin and corynoline was due to the affection of the first signal in the activation of the NLRP3 inflammasome.

On the other hand, cells were treated with 10 μM wogonin and 1 μM corynoline, respectively, for 1 h and then stimulated with LPS for a further 4 h. WB showed that wogonin and corynoline could inhibit the phosphorylation of NF-κB p65, the expression of NLRP3 and pro-IL-1β protein, which led to the reduction of downstream pro-inflammatory factors IL-1β release (Figure F).

Discussion

As a commercial anti-inflammatory oral prescription of TCM, PDL has been widely applied in treating anti-inflammatory diseases, especially upper respiratory infection diseases.[30] It has the functions of reducing fever and detoxifying, anti-inflammatory, and antiswelling. In order to better provide the guidance on clinical medication, we evaluated the anti-inflammatory effects of PDL on several classic inflammatory models. The results showed that PDL has a significant inhibitory effect on various types of inflammation. For instance, in an acute pharyngitis model, represents an upper respiratory infection, PDL decreased the levels of pro-inflammatory cytokines IL-6 and IL-1β in the peripheral blood and reduced inflammatory infiltration of pharyngeal tissues. In ALI model, PDL could reduce the degree of oedematous lung tissues, decrease the number of neutrophils, and inhibit the release of typical inflammatory cytokines IL-6, IL-1β, and TNF-α, which suggested PDL could inhibit the lower respiratory tract inflammation. To further investigate whether PDL can also resist nonspecific inflammation, we established three classic animal models, including xylene-induced ear swelling, carrageenan-induced acute paw edema, and capillary permeability-increasing models. Our results showed that PDL could significantly alleviate the ear swelling and reduce the inflammatory infiltration in mice ear tissues, inhibit the paw edema rate of rats, and decrease the capillary permeability in mice. These data demonstrated that PDL has a potent anti-inflammatory efficacy on both respiratory tract inflammation and nonspecific inflammation.

The potent anti-inflammatory efficacy of PDL has been verified in both clinical application and in vivo experiments. However, a systemic dissection of its underlying mechanisms is still stagnant because of the complex composition and multiple targets of PDL. Because of the holistic conception and complex composing principle of TCM, the existing methods widely used in studying western medicines are not scientifically eligible to explore the pharmacological mechanisms of herbal formulae.[31] Under these circumstances, the development of systems pharmacology (also known as network pharmacology) provides us with a novel and efficient approach to predict the active ingredients and their corresponding molecular targets. It is of significance to facilitate the modernization of TCM.

In this study, we utilized computational systems pharmacology methods for bioactive ingredients screening, target fishing, network establishment, and enrichment analysis to predict the mechanisms of PDL against inflammation. Subsequently, we verified the predictive findings with an acute pharyngitis model in vivo. As a result, 370 chemical ingredients were retrieved based on the four herbal medicines in PDL from the TCMSP database, among which 45 components were predicted to possess bioactive efficacy. Taken baicalein and norwogonin in S. baicalensis Georgi,[32,33] eupatorine and liquiritigenin in I. tinctoria L.,[34,35] dihydosanguinarine and stylepine in C. bungeana Turcz,[36,37] and quercetin and luteolin in T. mongolicum Hand.-Mazz(38,39) as examples, several chemical ingredients in these 45 compounds have been reported to exhibit potent anti-inflammatory activities. These data, at least, partly demonstrated the rationality and scientificity of the methodology of a network pharmacology approach.

Furthermore, we identified 38 inflammation-associated genes targeted by PDL via overlapping the 185 potential target genes into 471 inflammation-associated genes retrieved from three databases, as mentioned above. To decipher the underlying molecular mechanisms of PDL on anti-inflammation, we established an H–C-T-D network with 88 nodes and 290 edges using Cytoscape 3.7.1 software. Moreover, we performed enrichment analysis on the potential target genes to better understand the main biological processes that PDL exerted its anti-inflammatory efficacy. According to the p-value analyzed by the DAVID Bioinformatics Resources 6.8 System, of all the pathways involved, HIF-1 and TNF pathways were ranked the top two signaling pathways.

HIF-1 is a pivotal regulator of the transcriptional response to hypoxia, a notable feature of disorders where tissues suffer from chronic inflammation.[40] HIF-1 signaling is involved in multiple diseases, including arthritis, inflammatory bowel disease (IBD), asthma, cancer, and so forth.[41,42] TNF is a well-recognized cytokine with crucial functions in maintaining homeostasis and mediating diseases like rheumatoid arthritis, IBD, ankylosing spondylitis, as it plays a crucial role in the process of inflammation, apoptosis, and necroptosis.[43−45] Interestingly, PI3K and p38 are the mutual and critical genes involved in both HIF-1- and TNF-mediated signaling pathways. As a family of lipid kinases sharing a core motif, PI3Ks promote the production of pro-inflammatory cytokines like IL-6 by activating NK-κB[46,47] as we found in the acute pharyngitis rat model, and prominently the phosphorylation of PI3K and p38 in β-hemolytic streptococcus-induced acute pharyngitis. The serum levels of IL-6 and IL-1β significantly increased in accordance with the enhanced p-PI3K and p-p38 levels. Therefore, PDL possessed remarkable efficacy in ameliorating the inflammatory responses of acute pharyngitis, which might be highly correlated with deactivating PI3K and p38 signaling instead of inhibiting the total protein expression and further decrease the production of pro-inflammatory cytokines in the peripheral blood. Taken together, our data confirmed the therapeutic targets of PDL predicted by computational prediction were indeed involved in anti-inflammatory mechanisms in vivo.

Moreover, the systems pharmacology study showed that quercetin, wogonin, luteolin, baicalein, corynoline were the active components with the high content in PDL. Previous studies have demonstrated that baicalin[48,49] or wogonoside,[50] which have high contents in PDL, affect anti-inflammation by modulating the NLRP3 inflammasome signaling. The contents of wogonin and corynoline are also high in PDL, which rank closely followed by baicalin and wogonoside.[20] Few of the studies that reported wogonin and corynoline, might play an essential role in the anti-inflammatory process. Therefore, we hypothesized that similar to baicalin and wogonoside, wogonin and corynoline might alleviate inflammation also by suppressing NLRP3 inflammasome activation. This study found that wogonin and corynoline could not suppress the release of IL-1β after activation of the first stimulation signal. However, adding compounds before LPS can effectively inhibit IL-1β release. It suggested that wogonin and corynoline inhibit the first stimulation signal of inflammasome activation against inflammation. The WB assay showed that wogonin and corynoline reduced the activation of the NF-κB, inhibited the expression of the downstream protein involved in NLRP3 inflammasome activation, including NLRP3 and pro-IL-1β. It indicated that wogonin and corynoline inhibited the first signal of NLRP3 inflammasome activation to reduce the release of IL-1β. Actually, studies[51] have shown that baicalin, baicalein, and wogonin,[52] the main components of PDL, possessed anti-inflammatory effects previously. The anti-inflammatory activity is mainly due to their abilities to scavenge the reactive oxygen species and the improvement of antioxidant status by attenuating the activity of NF-κB and suppressing the expression of several inflammatory cytokines and chemokines, including monocyte chemotactic protein-1 (MCP-1), cyclooxygenases, lipoxygenases, cellular adhesion molecules, TNF, and interleukins.[53] Nevertheless, the pharmacological mechanisms of PDL on anti-inflammation still needs to be further studied, and more therapeutical targets are expected to be explored.

Collectively, PDL has a potent anti-inflammatory efficacy in both respiratory tract inflammation and nonspecific inflammation animal models in vivo. The approach of systems pharmacology successfully helped us screen out the bioactive components and predicted the potential therapeutic targets of PDL, HIF-1- and TNF-mediated signaling, which provided valid evidence for us to verify the underlying molecular mechanisms of PDL. We found that PDL downregulated HIF-1-and TNF-mediated signaling pathways through inhibiting PI3K and p38 phosphorylation in vitro. The bioactive components of PDL including wogonin and corynoline exerted anti-inflammatory potency via inhibiting NLRP3 inflammasome activation.

Conclusions

The combination of systems pharmacology and further verification experiments preliminarily provided the evidence that several bioactive ingredients in PDL synergistically exerted anti-inflammation efficacy by regulating HIF-1 and TNF pathways and the NLRP3 inflammasome. Besides, our study further confirmed that systems pharmacology could significantly contribute to the comprehensive understanding of the material basis and mechanisms of actions of TCM.

Materials and Methods

Reagents

PDL was provided by Jichuan Pharmaceutical Co., Ltd (1807012; Jiangsu, China). Dexamethasone (DEX) was from Shanghai ShangYao Xinyi Pharmaceutical Co., Ltd (015180207, Shanghai, China). LPS was purchased from Sigma (35H4086, MO, USA). ASA was from Bayer Health Care Manufacturing S.r.l (BJ41873, Leverkusen, Germany). Corynoline (CAS: 18797-79-0, ID: AQSY-0DVP) and wogonin (CAS: 632-85-9, ID: HAS8-PMVN) were purchased from the China National Institutes for Food and Drug Control (Beijing, China). Amoxicillin (AMX) was purchased from Shanxi Tongda Pharmaceutical Co. (181202, Shanxi, China).

Animals

ICR mice (18–22 g) were purchased from the Nanjing Qinglongshan animal breeding farm (Jiangsu, China, SCXK(Su) 2017-0001) and Zhejiang Vital River Laboratory Animal Technology Co. Ltd. (Zhejiang, China, SCXK(Zhe) 2018-0001). SD rats (180–220 g) were purchased from Shanghai Jiesijie Laboratory Animal Technology Co. Ltd. (Shanghai, China, SCXK(Hu) 2018-0004) and Beijing Vital River Laboratory Animal Technology Co. Ltd., (Beijing, China). All animals were raised at Nanjing University of Chinese Medicine under specific pathogen-free conditions at 22–25 °C and 40–65% humidity. All procedures involving animals were approved by the Animal Care and Use Committee of Nanjing University of Chinese Medicine and were performed strictly according to the Guide for the Care and Use of Laboratory Animals.

Acute Pharyngitis Model Establishment and Administration

Sixty SD rats, half male and half female, were randomly allocated into six groups including control, model, AMX at 0.18 g/kg, PDL at 0.9 g/kg, PDL at 2.7 g/kg, and PDL at 8.1 g/kg groups. Administration was performed from day 1 to day 7, and mice in the control and model groups were given an equal volume of normal saline once a day. The acute pharyngitis rat model was established by injecting 4 × 107 CFU β-hemolytic streptococcus (identification number: 32210; obtained from the National Institutes for Food and Drug Control, NIFDC, Beijing, China) in 0.1 mL of normal saline into the pharyngeal mucosa on day 5 and day 6. On day 7, mice were euthanized, and subsequently, the pharyngeal mucosa and peripheral blood samples were collected 24 h after the last injection.[54]

ALI Model Establishment and Administration[55]

Mice were given PDL (1.3, 3.9, and 11.7 g/kg) and DEX (5 mg/kg) for 7 days by intragastric administration. Except for the control group, each group was treated with LPS physiological saline solution (5 mg/kg) for mouse airway instillation to replicate the ALI model.

Acute Ear Swelling Model Establishment and Administration[56]

Mice were given PDL (1.3, 3.9, and 11.7 g/kg) and ASA (0.2 g/kg) for 7 days by intragastric administration. Except for the control group, the other groups used 30 μL of xylene to smear the mouse auricle to reproduce the acute ear swelling model. The thickness of mouse ears was measured using a thickness gauge, and the ear tissue was microscopically examined after HE staining.

Acute Paw Edema Model Establishment and Administration

Rats were given PDL (0.9, 2.7, and 8.1 g/kg) and ASA (0.2 g/kg) for 7 days by intragastric administration. Except for the control group, the rats in the other groups replicated the acute paw edema model by subcutaneous carrageenan injection.[57] The paw volume of the rats was measured every hour. Swelling rate (%) = (foot volume after inflammatory–foot volume before inflammatory)/foot volume before inflammatory by 100%.

Capillary Permeability Model Establishment and Administration

Mice were given PDL (1.3, 3.9, and 11.7 g/kg) and ASA (0.2 g/kg) for 7 days by intragastric administration. Mice were injected with 2% Evans blue physiological saline solution in the tail vein. Except for the control group, mice in each group were subsequently injected intraperitoneally with 0.6% acetic acid saline solution to replicate the capillary permeability-increasing model. The capillary permeability was determined according to the content of Evans blue in the abdominal fluid of mice.[58]

Cell Culture

Primary peritoneal macrophages cells were extracted from mice as described previously[59] and seeded into 96-well plates at a density of 5 × 105 per well, cultured with 1640 medium containing 10% fetal bovine serum and 1% penicillin/streptomycin. Cells were pretreated with wogonin (0.1, 1, 10, and 100 μM), corynoline (0.1, 1, 10, and 100 μM), or medium with 0.05% DMSO as the vehicle (control) for 24 h, then test the viability using the MTT assay.

Peritoneal macrophages cells were cultured under the same condition as above. Cells were treated with wogonin or corynoline before or after adding LPS. ELISA detected the level of IL-1β, the expressions of NLRP3, NF-κB p65, p-P65, and pro-IL-1β were detected by WB.

Lung Dry/Wet (W/D) Determination

Mice were anesthetized by intraperitoneal injection of 50 mg/kg pentobarbital sodium, before the rats were euthanized using CO2 as previously described. Mice were sacrificed to obtain lung tissues, which were drained and measured for wet weight. Afterward, lung tissues were incubated at 80 °C for 4 h, and then tissues were weighed to obtain measures of dry weight.

MPO

The measures of the activity of MPO as the instructions of the kit described. MPO activity (U/tissue wet weight) = (measured OD value – control OD value)/11.3 × sample volume (g).

Neutrophils Count

The total number of cells in BALF was counted, 0.02 mL was taken for smear and Regeisen’s staining, and cell classification and counting were performed under a microscope.

Determination of IL-6, IL-1β, and TNF-α Production

Production of IL-6 (4300713, eBioscience, CA, USA), IL-1β (B288648, Biolegend, CA, USA), and TNF-α (B288663, Biolegend, CA, USA) in BALF were measured using ELISAs according to the manufacturer’s instructions. Total protein levels in the homogenates were examined using a bicinchoninic acid (BCA) kit (041318180603, Beyotime, Shanghai, China). IL-6, IL-1β, and TNF-α protein levels were assessed with the formula: concentration of IL-6, IL-1β, and TNF-α in the homogenate/total protein (pg/mg).

Histological Analysis

Histological analysis was performed as described previously.[60] Pharyngeal mucosa was inflated with 4% paraformaldehyde and fixed in formalin. Afterward, pharyngeal mucosa was embedded in paraffin and dissected at a thickness of 5 μm for HE staining. Sections were then stained according to a standard protocol. Stained sections were analyzed by pathologist using single-blind method. The following observations were recorded: (1) the degeneration and necrosis of epithelial cells, and the inflammatory cells infiltration in or under the epithelium; (2) the exudate within the pharynx cavity; and (3) the hyperemia and dropsy of the pharyngeal wall or inflammatory cell infiltration within the pharyngeal wall. The degree of inflammation was evaluated semiquantitatively using the scores of 0–3 to indicate no, mild, moderate, and severe inflammation.

Western Blotting

WBs were performed as described previously.[60] Tissue homogenates and cells were prepared in lysis buffer (Beyotime, Shanghai, China), consisting of 1 nM phenylmethanesulfonyl fluoride (Beyotime, Shanghai, China) to extract protein, and the protein concentration was detected using an BCA protein detection kit. An equal amount of protein solubilization was added to the band and separated by 10% sodium dodecyl sulfate polyacrylamide gel. Then the isolated protein was transferred to a polyvinylidene fluoride membrane (Millipore Corporation, Darmstadt, GER). After incubation in 5% skimmed milk containing 0.1% TBST for 1 h, the membrane was incubated with primary antibodies against mouse PI3K(A1520, Santa Cruz Biotechnology, CA, USA), phospho-PI3K(D1718, Santa Cruz Biotechnology, CA, USA), p38(C0218, Santa Cruz Biotechnology, CA, USA), and phospho-p38 antibody (I1719, Santa Cruz Biotechnology, CA, USA), NLRP3(A27381510, Adipogen, Liestal, SUI), NF-κB p65(#F2912, Santa Cruz Biotechnology, California, USA), p-NF-κB P65 (16, Cell Signaling Technology, Boston, MA, USA), and Pro-IL-1β (#12242, Cell Signaling Technology, Boston, MA, USA) or GAPDH (BC004109, Proteintech, PA, USA) in 1:1000 dilution overnight at 4 °C. Then, the secondary antibodies were added, and the membrane was incubated at room temperature for 1 h. In each sample, the target protein expression level was normalized to GAPDH.

Systems Pharmacology

The active compounds of PDL were obtained from the TCMSP database (http://tcmspw.com/index.php),[61] a unique system pharmacology platform designed for herbal medicines. Then, four in silico ADME models, including human oral bioavailability (OB), drug-likeness (DL), lipophilic prediction (A log P), and small intestinal epithelial cell permeability prediction (Caco-2) were employed to explore the candidate compounds in PDL.[62] The threshold values for these screening models were set to OB ≥ 30%, DL ≥ 0.18, A log P ≥ 5, and Caco-2 ≥ −0.4, respectively.[63−66] The compounds which satisfy all the criteria are listed as candidate molecules. We downloaded the 3D structures of the screened active ingredients in PDL on PubChem (https://pubchem.ncbi.nlm.nih.gov/) or on SciFinder (https://scifinder.cas.org/) and then uploaded them to PharmMapper (http://www.lilab-ecust.cn/pharmmapper/index.html), a pharmacophore matching and potential target identification platform. The active ingredient performs reverse docking in the server to fish the potential targets corresponding to the chemical components. Then, we collected inflammatory targets from three sources. One was the Therapeutic Target Database (TTD, http://db.idrblab.net/ttd/), a database that provides information about the known and explored therapeutic protein and nucleic acid targets, the targeted disease, pathway information, and the corresponding drugs directed at each of these targets. Another resource was the DrugBank database (https://www.drugbank.ca, version 4.3), a unique bioinformatics and cheminformatics resource that combines detailed drug data with comprehensive drug target information. The other one is DisGeNET (https://www.disgenet.org), a discovery platform containing one of the largest publicly available collections of genes and variants associated with human diseases. We selected the keyword “inflammation” used for drugs approved by the Food and Drug Administration (FDA) to treat inflammation and human gene/protein targets. The targets corresponding to the potential ingredients in PDL and the targets corresponding to inflammation were taken as the intersections. The establishment of the C–T–D network using Cytoscape 3.7.1 software could help identify each compound’s protein targets and understand the mechanism of action of multicomponents and multitargets in TCM. The predicted targets of PDL were analyzed by the KEGG pathway using the DAVID Bioinformatics Resources 6.8 System (https://david.ncifcrf.gov/),[67] and p-value ≤ 0.05 was considered to have a significant enrichment effect.

Statistical Analysis

Data are expressed as the mean ± SD. One-way ANOVA analysis was used for multiple group comparisons using GraphPad Prism 6 (GraphPad Software, San Diego, CA, USA). Values of P < 0.05 indicated statistical significance.