The effect and mechanism of Modied Guo-Min Decoction and Yu-Ping-Feng Powder on ne particulate matter-induced lung inammatory injury and airway mucus hypersecretion in rats

Exposure to ne particulate matter (PM2.5) severely impairs public health. The mechanism of PM2.5-induced lung injury is complex and diverse. Modied Guo-Min Decoction (MGMD) and Yu-Ping-Feng Powder (YPFP) have been found to improve clinical symptoms in respiratory patients during smog weather, but the mechanism remains unclear. This study aimed to investigate the effect and mechanism of YPFP and MGMD against PM2.5-induced lung injury. of cells cytokines, phosphorylation


Background
According to the latest World Health Statistics published in 2019, around 91% population of the world lived in the circumstance where the air quality inferior to WHO's suggestion, and an estimated 4.2 million people died due to ambient air pollution. Fine particulate matter (PM2.5) which less than 2.5 microns in diameter is the major source of air pollutant [1]. The average annual concentration of PM2.5 in Chinese urban area in 2016 (51 µg/m 3 ) was far beyond the level WHO recommended (12 µg/m³). Recent epidemiological studies have shown that both long and short term exposure to PM2.5 are closely associated with respiratory diseases which increased the risk of chronic obstructive pulmonary disease (COPD), lung cancer, asthma and pneumonia [2][3][4]. The mechanisms of PM2.5-induced respiratory system injury have been investigated including in ammatory response, oxidative stress, genotoxicity, and imbalanced intracellular calcium homeostasis [5,6].
The in ammatory response which runs through the process of several respiratory diseases is the major cause of PM2.5-induced lung injury. PM2.5 destroys the epithelium and initiates the in ammatory response by stimulating epithelium cells and pulmonary macrophage to express cytokines, e.g., the interleukin (IL)-1β, interleukin (IL)-6, interleukin (IL)-17A, tumor necrosis factor (TNF)-α, and Monocyte chemoattractant protein (MCP)-1 [7]. These in ammatory factors can activate in ammatory cells in ltration and attraction such as neutrophils, mononuclear macrophages, and lymphocytes which will aggravate in ammatory injury and immune function in turn [8]. The TLR2-MyD88-NFκB signalling pathway is closely associated with in ammatory response and plays a vital role in multiple respiratory diseases [9]. Toll-like receptor (TLR) 2 is a transmembrane protein which is a critical component of detecting pathogen and initiating immunity in the respiratory epithelium [10,11]. After activated by in ammasomes, TLR2 induces the recruitment of myeloid differentiation primary response gene 88 (MyD88) by ligand binding. Consequently, the downstream nuclear factor kappa B (NFκB) is phosphorylated. Typical NFκB structure is the heterodimer consisting of two subunits, RelA (p65) and p50. After activated, NFκB translocates into the nucleus, leading to the expression of in ammatory mediators [12]. NFκB is not only a part of in ammatory processes but an important role in tumorigenic function. The main target genes of TLR2-MyD88-NFκB signalling pathway include IL-1, IL-6, IL-8, TNF-α, MCP-1, vascular cell adhesion molecule 1(VCAM-1) and the intercellular cell adhesion molecule 1 (ICAM-1). Recent studies showed that PM2.5 could lead to exacerbation of airway in ammation by activating TLR2-NFκB signalling pathway [13]. Type 2 in ammation which involves eosinophils, mast cells, basophils, T H 2 cells is another important mechanism of respiratory diseases, such as asthma, bronchial hyperresponsiveness, and allergic rhinitis. These immune cells secret type 2 cytokines (e.g., IL4, IL5 and IL13), IgE, and, bioactive substance which is essential for airway remolding and mucus hypersecretion. [14]. A study from Japan discovered that exposure to PM2.5 could cause severe eosinophil in ltration in airways and increase the level of IL-4, IL-5, and IL-13 in bronchoalveolar lavage uid (BALF) of BALB/c mice [15].
Mucus hypersecretion is both the pathogenic factor and pathological product of various airway diseases including asthma, chronic obstructive pulmonary disease, and cystic brosis. Airway mucus which major component is mucin5AC (MUC5AC) is mainly secreted by goblet cells in the bronchial epithelium [16]. It has been investigated that PM2.5 could induce MUC5AC expression in the rat airway after 3 months' exposure [17]. MUC5AC synthesis and release can be initiated by several signalling pathway especially EGFR-PI3K-AKT. Stimuli such as PM2.5 and in ammatory mediator activate epidermal growth factor receptor (EGFR) across the cell membrane. These transduction signals motivate the downstream phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) and protein kinase B (AKT) to phosphorylate which induces MUC5AC mRNA expression in nuclear [18]. A research in vitro showed that PM2.5 induced mucus hypersecretion by activating EGFR-AKT pathway in human bronchial epithelial cell (HBEC) [19].
Modi ed Guo-Min Decoction (MGMD) is transformed from Guo-Min decoction (GMD) which is an empirical formula designed by TCM physician Zhu Chenyu. GMD is consist of Radix Stellariae (Yinchaihu), Radix Saposhnikoviae (Fangfeng), Fructus Mume (Wumei), and Fructus Schisandrae Chinensis (Wuweizi). This formula is mainly applied in treatments for allergic diseases such as asthma, cough variant asthma (CVA), allergic rhinitis, and atopic dermatitis. The mechanism might be associated with downregulating IL-4 and IL-5 mRNA expression in lung tissues [20]. In clinical therapy, we found that GMD with the addition of Radix Peucedani (Qianhu) and Radix Platycodonis (Jiegeng) was effective in patients with PM2.5-induced cough and phlegm.
Yu-Ping-Feng Powder (YPFP) which consist of Radix Astragali (Huangqi), Radix Saposhnikoviae (Fangfeng), and Rhizoma Atractylodis Macrocephalae (Baizhu) is a classic Chinese medicine formula recorded in Jiu Yuan Fang. YPFP was widely used in respiratory diseases for the main mechanism that it could relieve in ammation and promote transformation of immunological cells [21,22]. In the clinical application, we found that YPFP could alleviate PM2.5-induced respiratory symptoms. However, the underlying mechanisms of both two formulas on PM2.5-related pulmonary injury remain unclear. In this study, we investigated and compared the effects of MGMD and YPFP on PM2.5-induced lung injury of Wistar rat and their mechanisms.

PM2.5 suspension preparation
The PM2.5 collecting site was on the roof of Chinese Research Academy of Environmental Sciences (Beijing, China). The sample was collected on Te on lters (Pall Corporation, New York, USA) 8 h/day for consecutive 2 weeks in January 2018 by using Thermo Anderson sampler (PDR-1500, Thermo Fisher Scienti c, Inc., Waltham, MA, USA). The lter membrane was then cut into 2 × 2 cm sections and immersed in distilled water for ultrasonic oscillation at room temperature for 30 min x 3 times. After ltrated with gauze, the solution was dried overnight in a freezing vacuum machine (LABCONCO, Kansas City, MO, USA). The particles were weighed and resuspended in normal saline (NS) to a nal concentration 0.65 g/mL.

Animal and experimental protocol
Twenty Wistar rats (male, 6-8 weeks old, weighing180-220 g) were obtained from Vital River Laboratory Animal Technology Co.,Ltd. (Beijing, China). All rats were kept in a 12 h dark/light cycle and constant temperature (25℃) room with enough food and sterile water. The experimental protocol was approved by the Experimental Animal Ethics Committees of Peking University People's Hospital (No. 2017PHC022).
The rats were randomly separated into four groups which were Control group (n = 5), PM2.5 group (n = 5), MGMD group (n = 5) and YPFP group (n = 5). Every three days from the rst day of the experiment, PM2.5 group, MGMD group and YPFP group were intratracheal instilled 0.1 mL PM2.5 suspensions via rats' throat using a 18G remaining catheter after anesthesia with 5% iso urane, while Control group was instilled 0.1 mL NS.

Sample harvesting
On day 29, after anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg⋅bw) all rats were sacri ced by blood collection from abdominal aorta. Opened rat's chest, tied up right lung hilum and isolated right lung. The middle lobe was xed by 4% paraformaldehyde for 48 h, dehydrated in a gradient ethanol series, and embedded in para n for morphology detection. The anterior and posterior lobes of right lung were stored in liquid nitrogen for qPCR and Western Blot analysis. 5 mL PBS infused into right lung and pumped 3 times to collect bronchoalveolar lavage uid (BALF). The liquid was then centrifuged at 3000 rpm for 10 min at 4℃. The supernatant was aspirated and maintained at -80℃ for ELISA analysis.

HE and AB/PAS-Staining
The para n blocks were cut into 4 µm sections. The sections were sequentially depara nized, rehydrated and stained with hematoxylin and eosin (H&E). After dehydrated and mounted, the sections were nally observed under light microscope (Leica DFC400, Germany).
Alcian Blue/Periodic Acid-Schiff (AB/PAS) staining was performed to detect goblet cell metaplasia of bronchial epithelium. The images of lung tissues were captured by microscope. AB/PAS positive area and total area of corresponding bronchial epithelium were measured by the software Image-Pro Plus 6.0. Data were presented as the ratio of AB/PAS-positive area to the total area.

Statistical analysis
The result was expressed as mean ± SD. We use SPSS 22.0 software (IBM; Armonk, NY, USA) to determine the statistical differences between groups by One-way analysis of variance (ANOVA). P < 0.05 was set as statistically signi cant.

Effect of MGMD and YPFP on PM2.5-induced Pulmonary Histopathological Injury
In Control group, the airway epithelial cells with intact cilia arranged regularly. In lamina propria of normal rats, there were few in ammatory cells. It could be hardly found hyperplastic gland around the airway. In the visual eld of alveoli, the structure remained integrity without alveolar fusion, collapse, interstitial edema, or hyperemia. Compared with Control group, lung of PM2.5 group showed obvious histological injuries such as thickened and swollen epithelium, adhesion and loss of cilia, gland hyperplasia, alveolar fusion and collapse, interstitial edema and congestion, and in ammatory cells in ltration. In both MGMD and YPFP groups, the epithelial injures and alveoli destruction were signi cantly inhibited compared with PM2.5 group (Fig. 1A). As shown in Fig. 1B, airway smooth muscle in PM2.5 group and YPFP group has the hypertrophy trend compared with Control group while the area of airway smooth muscle in MGMD group tended less than that in PM2.5 group. However, there was no statistical signi cance between each group. In terms of eosinophils around the airway, there was a remarkable increase in PM2.5 group compared to those in Control group (0.93 ± 0.55 vs 9.00 ± 0.85, P < 0.01). Mean counts of eosinophils in both MGMD and YPFP groups have noticeable decrease compared with PM2.5 group (4.07 ± 2.70 vs 9.00 ± 0.85, P < 0.01 and 4.00 ± 0.78 vs 9.00 ± 0.85, P < 0.01, Fig. 1C).

Effect of MGMD and YPFP on PM2.5-induced In ammatory Cells In ltration and Mediators Expression
The effects of MGMD and YPFP on the lung of PM2.5 exposed rats were determined by molecular biology experiments on the in ammatory cells surrounding the airway, cytokines in BALF, and in ammatory mediators in lung tissues. We observed that prolonged exposure to PM2.5 could increase the in ltration of the in ammatory cell around the airway. MPO, CD68, CD4, and CD8 were the speci c markers of activated neutrophil, macrophage, and T lymphocytes, respectively. As shown in Fig. 2, compared with Control group, multiple in ammatory cells in the submucosa and lamina propria of the airway were signi cantly increased in PM2.5 group. After being intervened by MGMD and YPFP, the in ltration was restrained to a certain extent.
Furthermore, we detected FcεRI which played an essential role in IgE releases in the lungs of each group. The data delivered that the level of FcεRI in lung tended upwards after exposure to PM2.5 with no signi cance. MGMD and YPFP had the inhibitory effect on FcεRI expression to a certain degree (P < 0.01 and P < 0.01) (Fig. 5A and E).

Effect of MGMD and YPFP on PM2.5-induced goblet cell metaplasia and mucus hypersecretion
Goblet cells in airway epithelium were stained bluish-violet by AB/PAS-staining. The images were captured under the light microscopy and the count of goblet cells were obtained by IPP6.0. The quantitation of goblet cell metaplasia was expressed as the ratio of positive staining area to epithelium area. As shown in Fig. 6A, it was hardly found goblet cell in Control group whereas marked hyperplasia was in PM2.5 group. Goblet cell metaplasia could also be found in MGMD group and YPFP group, but exhibited decreasing trend compared to PM2.5 group. Besides, YPFP showed more effective against goblet cell hyperplasia than MGMD. The proportion of goblet cell to epithelium were 0.35 ± 0.36%, 25.56 ± 2.58%, 18.17 ± 3.45% and 12.72 ± 4.06% in Control group, PM2.5 group, MGMD group, and YPFP group, respectively. There was a noticeable increase of average area of positive stains in PM2.5 group compared with Control group (P < 0.01). Both MGMD and YPFP could inhibit goblet cell metaplasia (P < 0.01, P < 0.01), while YPFP was more potent (P < 0.05) (Fig. 6B).
MUC5AC, the main component of mucus in airway, was detected by IHC staining, Western Blot and qPCR respectively in each group. As displayed in Fig. 7A, the IHC positive staining of MUC5AC was the tawny area among epithelium. MUC5AC could be rarely detected in normal epithelial lamina, while PM2.5 group resulted in signi cant increase. Compared with the model group, MGMD and YPFP could inhibit MUC5AC expression. The quantitation presented as the percentage of staining area among airway epithelium proved the statistical signi cance as the images showed. But only YPFP group manifested statistical signi cance (Fig. 7B). qPCR for MUC5AC mRNA presented the similar results (Fig. 7C). MUC5AC protein expression in lung tissues analyzed by Western Blot analysis indicated that PM2.5 exposures could lead to MUC5AC overexpression. Two intervention groups showed the inhibitory effects with obvious signi cance (Fig. 7D and E).

Effect of MGMD and YPFP on TLR2-MyD88-NFkB and EGFR-PI3K-AKT Signalling Pathway
To further investigate the mechanism of the inhibitory effects of MGMD and YPFP on in ammation and mucus hypersecretion, we applied the Western Blot analysis of the lung tissues in each group. As displayed in Fig. 8A-D, the level of TLR2 and MyD88 proteins and NFκB phosphorylation remained the same in four groups. In terms of the EGFR-PI3K-AKT signalling pathway, results showed that there were remarkable increases in phospho-PI3K (P < 0.01) and phospho-AKT (P < 0.05) of PM2.5 group compared with Control group. The phosphorylation of PI3K (P < 0.05) and AKT (P < 0.05) was decreased to varying degrees in YPFP group with statistical difference, while MGMD only showed effective on phospho-AKT (P < 0.05) (Fig. 8E-G).

Discussion
According to recent clinical studies, both long and short period exposure to PM2.5 could increase the prevalence and mortality of respiratory diseases [2][3][4]. In our clinical application, we discovered that MGMD and YPFP could relieve the respiratory symptoms caused by hazy weather. However, the mechanism is still unclear. We designed this research to explore the effect and underlying mechanism of MGMD and YPFP against the PM2.5-induced lung injury. We established an animal model of PM2.5induced lung injury by intratracheal instillation of PM2.5 suspensions and treated with MGMD and YPFP.
The results of HE staining demonstrated that PM2.5 could induce the damage of airway epithelium, destruction of pulmonary alveoli, and in ammatory cell in ltration in lung tissues of rats. After treatment with MGMD or YPFP, we found that lung injury could be attenuated to a certain extent.
To further investigate the types of in ammatory cells, we applied immunohistochemistry and discovered that PM2.5 signi cantly increased neutrophil, macrophage, and T lymphocyte in ltration surrounding the airway, which the speci c markers MPO, CD68, CD4, and CD8 were stained positive respectively. The in ammation can be staged into acute in ammation, subacute in ammation, and chronic in ammation [23]. Acute in ammation is characterized by neutrophils and monocyte in ltration, which initiated by rolling along and sticking to the epithelium. After been stimulated by PM2.5, the airway epithelial cells and alveolar macrophages secreted adhesion molecule, chemokines, and cytokines. In this study, we detected that PM2.5 could dramatically increase ICAM-1 expression, especially in airway epithelium by morphology. We achieved the same result in the protein level of total lung tissues. ICAM-1 can lead to leukocyte adhesion and migration into airway submucosa and alveolar interstitium by binding to the ligands (e.g., LFA-1, Mac-1, integrins, and brinogen). Moreover, the overexpression of ICAM-1 may increase the risk of common colds, for it is also the binding site for rhinovirus [24]. MCP-1, primarily expressed by macrophages and epithelial cells is another in ammatory mediator recruiting monocytes. In the early stage of exposure to PM2.5, MCP-1 elevated in the lung tissue and BALF [25]. The current study demonstrated that PM2.5 upregulated the MCP-1 expression in the airway epithelium and lung tissues. The same trends of IL-1, TNF-α, and IL-17A, the typical pro-in ammatory cytokines, could be found in BALF. The overexpression of NE in the lung tissues of model rats also indicated the augment and activation of neutrophils. This evidence demonstrated that PM2.5 could initiate acute in ammation. Except for the acute in ammatory cells, we also detected macrophages and lymphocytes in ltration and gland proliferation the characteristic in chronic in ammation. Therefore, PM2.5-induced in ammatory response was the result of the combination of acute and chronic in ammation.
Both MGMD and YPFP could reduce neutrophil, monocyte, and lymphocytes in ltration for the expression of MPO, CD68, CD4, and CD8 were obviously inhibited. In terms of pro-in ammatory mediators, the results of IHC and Western Blot demonstrated that MGMD and YPFP effectively downregulated the expression of MCP-1 in lung tissues. In the study of ICAM-1, another in ammatory medium, YPFP remarkably reduced the over-expression of ICAM-1 stimulated by PM2.5, but no signi cant inhibitory effect was found in MGMD group. Furthermore, the levels of IL-1β and IL-17A in BALF and NE in lung tissue showed that YPFP was more potent in inhibiting lung tissue in ammation than MGMD.
In addition to type 1 in ammation, HE-staining showed that there was a mass of eosinophils in model rats' lung tissues, which was the symbol of type 2 in ammation. Furthermore, IgE and IL-4 closely related to type 2 in ammation exhibited higher levels in BALF after treated with PM2.5 suspensions. These mediators are essential for airway remodeling and abnormal mucus secretion. Previous evidence indicated that PM2.5 exacerbated accumulation of eosinophils and neutrophils into the airway submucosa [26,27]. Another study reached the same conclusion that PM2.5 caused signi cant increase of eosinophils and neutrophils in BALF over time [28]. Therefore, we believed the PM2.5-induced in ammatory injury was the combined action of type 1 and type 2 in ammation. It could be found that both MGMD and YPFP had the inhibitory effects on PM2.5-induced eosinophil proliferation. The results indicated that MGMD and YPFP effectively decreased the levels of IL-4 and IgE in BALF, and the effect of MGMD on IL-4 was more potent. It was well acknowledged that the high a nity Fc receptor (FcεRI) played a mediating role in allergic in ammation. FcεRI mainly expressed on mast cell could bind to IgE and activate the release of histamine, prostaglandins, and leukotrienes which was known to exacerbate allergic response [29]. We detected that FcεRI protein in lung decreased markedly compared with model rats after MGMD and YPFP treatment.
It is considered that TLR2-MyD88-NFκB played an essential role in promoting in ammation. Several studies showed that PM2.5 could work as pathogen-associated molecular patterns (PAMP) combining with pattern recognition receptors (PRR) like TLR2 on the surface of the airway epithelial cells and alveolar macrophages. TLR2 delivered the activated signals to downstream MyD88 and then induced the phosphorylation of NFκB, which caused in ammatory cytokines synthesis. This signal transduction was generally recognized as an important mechanism of PM2.5-triggered in ammatory injury [26]. However, in the current study, the levels of TLR2, MyD88, and NFκB phosphorylation of each group are consistent with no signi cance. The reason for this is because, except for NFκB related pathway, there are numerous approaches promoting in ammation, such as MAPKs, Smads, and Nrf2 signalling pathways [30][31][32].
MUC5AC, the primary component of airway mucus, is synthesized and released by goblet cells in the epithelium. We detected by AB-PAS staining that exposure to PM2.5 could signi cantly exacerbate the goblet cell hyperplasia and MUC5AC release in the airway. The results of MUC5AC protein and mRNA expression in lung tissues came out the same trend. After treatment with MGMD and YPFP, we found that the therapy could reduce the goblet cell proliferation and inhibit MUC5AC expression. Moreover, the effects of YPFP was more signi cant than that of MGMD. In terms of the mechanism, it has been considered that EGFR-PI3K-AKT signalling pathway is tightly associated with regulating MUC5AC mRNA expression. PM2.5 and PM2.5-induced in ammatory mediators cause excess MUC5AC synthesis via the cascade of EGFR-PI3K-AKT signalling pathway [18]. Western Blot bands manifested remarkable increase in the ratio of phosphorylation of PI3K and AKT. However, phosphorylation of EGFR showed no difference in each group, indicating that EGFR is not the main signal receptor that causes the cascade of PI3K-AKT in this study. It has been investigated that receptor tyrosine kinase (RTK), G protein coupled receptor (GPCR) and T cell receptor can also bind to extracellular stimulation signals and activate the PI3K-AKT signalling pathway [33]. According to the results of Western Blot, YPFP had the effect of simultaneously inhibiting the phosphorylation of PI3K and AKT, whereas MGMD only showed a signi cant difference in AKT. After synthesized, MUC5AC protein was packed tightly in granule in the cytoplasm of goblet cell and remained stable until activated by stimuli like NE and MCP-1. MGMD and YPFP could downregulate the expression of NE and MCP-1, but IHC images showed that YPFP had more inhibitory effect on NE and thereby had advantage on suppressing MUC5AC release in turn.
According to TCM theory, the pathogenesis of PM2.5-induced respiratory symptom is "wind evil invading the lung". YPFP, a classic formula, can dispel wind-evil and consolidate super cies. In recent researches, YPFP was discovered to be effective in restricting in ammation, alleviating allergic response, repairing tight junction of epithelia barrier [34][35][36]. GMD which comes from experience prescription is mainly applied in various allergic diseases. Previous study suggested that the mechanism of GMD was associated with reducing the production of IgE and suppressing in ammatory cell in ltration [20]. In clinical application, GMD with Qianhu and Jiegeng was more potent in treating allergic symptom for the two herbs could recover the lung function of dispersing and descending. Recent researches also explored that Qianhu and Jiegeng could signi cantly attenuate eosinophilic airway in ammation, downregulate the expression of in ammatory cytokines, and inhibit the activity of Th2 cells [37,38].

Conclusions
In conclusion, this research indicated that MGMD and YPFP could signi cantly attenuate PM2.5-induced type1 and type2 in ammation of lung by regulating the in ammatory mediators. Meanwhile, both two therapies could inhibit the airway mucus hypersecretion through PI3K-AKT signalling pathway. The result suggested that YPFP had more advantage in preventing type1 in ammation and mucus hypersecretion, while MGMD was more bene cial in reducing type2 in ammation. However, more underlying mechanism remains to be explored further.  Effects of MGMD and YPFP on the histopathological change, airway smooth muscle and eosinophil of lung in PM2.5 rats. A: Control group displayed normal morphology of airway and alveoli. PM2.5 group showed airway and alveolar structural damage. The lung injury were mitigated in MGMD and YPFP (H&E staining, ×100 and ×400). B: Quantitated the airway smooth muscle by the ratio of smooth muscle area to airway area (n=5). C: The mean count of eosinophils around the airway in MGMD and YPFP groups Effects of MGMD and YPFP on the in ammatory cells in lung tissues of PM2.5 rats. A: MPO, the marker of activated neutrophils, was stained brown by IHC (magni cation ×100 and ×400). B: CD68, the surface marker of mononuclear macrophage, was stained by IHC staining (magni cation ×100 and ×400). C and D: T-lymphocytes which speci cally expressed CD4 and CD8 were detected by IHC staining respectively (magni cation ×100 and ×400).      Effect of MGMD and YPFP on the TLR2-MyD88-NFκB and EGFR-PI3K-AKT signalling pathway in lung tissues of PM2.5 rats evaluated by Western Blot. A: The bands of TLR2, MyD88, NFκB, p-NFκB, EGFR, p-EGFR, PI3K, p-PI3K, AKT, p-AKT, and β-actin as the internal control. B-G: Estimated the average optical density of each band. The statistics were expressed as target protein vs β-actin (mean ± SD, n=5).