The traditional herbal formulation, Jianpiyifei II, reduces pulmonary inflammation induced by influenza A virus and cigarette smoke in mice.

Chronic obstructive pulmonary disease (COPD) is a worldwide chronic inflammatory lung disease, and influenza A virus (IAV) infection is a common cause of acute exacerbations of COPD (AECOPD). Therefore, targeting viral infections represents a promising strategy to prevent the occurrence and development of inflammatory flare ups in AECOPD. Jianpiyifei II (JPYFII) is a traditional herbal medicine used in China to treat patients with COPD, and its clinical indications are not well understood. However, investigation of the anti-inflammatory effects and underlying mechanism using an animal model of smoking has been reported in a previous study by our group. In addition, some included herbs, such as Radix astragali and Radix aupleuri, were reported to exhibit anti-viral effects. Therefore, the aim of this study was to investigate whether JPYFII formulation relieved acute inflammation by clearing the IAV in a mouse model that was exposed to cigarette smoke experimentally. JPYFII formulation treatment during smoke exposure and IAV infection significantly reduced the number of cells observed in bronchoalveolar lavage fluid (BALF), expression of pro-inﬂammatory cytokines, chemokines, superoxide production, and viral load in IAV-infected and smoke-exposed mice. However, JPYFII formulation treatment during smoke exposure alone did not reduce the number of cells in BALF or the expression of Il-6, Tnf-a, and Il-1β. The results demonstrated that JPYFII formulation exerted an anti-viral effect and reduced the exacerbation of lung inflammation in cigarette smoke (CS)-exposed mice infected with IAV. Our results suggested that JPYFII formulation could potentially be used to treat patients with AECOPD associated with IAV infection.


Introduction
Chronic obstructive pulmonary disease (COPD) is a worldwide chronic inflammatory lung condition characterized by irreversible airflow limitation [1]. Acute exacerbation of COPD (AECOPD) is associated with a range of symptoms, including aggravated coughing, breathing difficulty, sputum production, and wheezing, and it is a leading cause of death in COPD patients [2]. The most recent GOLD report predicted that COPD would become the third highest cause of morbidity and mortality worldwide by 2030 [3]. In China COPD was the fourth leading cause of death in 2017 [4], and the mortality rate of in-hospital patients with AECOPD was 2.5-24.5% [5]. AECOPD accelerates declining lung function and induces numerous additional complications including heart failure, respiratory failure, pulmonary embolism, and pneumothorax, increasing the associated global economic burden annually [6][7][8].

IAV infection
After CS or sham exposure for four days, mice were anesthetized with isoflurane on day 5 and given 12 plaque-forming units (PFU) of A/PR/8/34 (H1N1) influenza virus (provided by Professor Zifeng Yang, Guangzhou Institute of Respiratory Health, Guangzhou, China) via nasal aspiration in 30 μl of minimal essential Eagle's medium (MEM) as previously described [16,31]. An equal volume of MEM was administered intranasally to vehicle-control mice. The mice were then culled 3 or 5 days after viral infection. Changes in body weight, an indicator of infection severity, were recorded daily.

JPYFII formulation preparation and treatment
JPYFII formulation is a combination of eight Chinese herbal medicines. JPYFII formulation was prepared as previous described [24]. In brief, eight certified dried herbal medicine were purchased from the Second Affiliated Hospital of Guangzhou University of Chinese Medicine (Table 1), powdered and extracted twice in ten times their volume of boiling water for 90 min. Each water extract was filtered and dehydrated under a vacuum. The residues were freeze-dried and stored in at 4 • C prior to use. The extract solution was freshly prepared daily by dissolving of 1.6 g freeze-dried powder into 1 ml of saline for 30 min and using ultrasonic dissociation to assist dissolving. Syringe filters with a pore size of 0.45 μm were used to remove any bacteria and insoluble powder before the mice were gavaged with the solution. The main compounds included in the formulation were analyzed and identified via liquid chromatography-mass spectrometry as previously described [24]. Oral gavage of JPYFII formulation (24 g/kg) was completed before 8 a.m. daily, which was 1 h before the first CS exposure or IAV infection. The mice received JPYFII formulation until they were assessed on day 5 post-infection. The JPYFII formulation dose used was based on results obtained from anti-inflammatory dose screening in an IAV-induced CS model (Supplementary Figure S1). It was determined that the 6.7 g/kg body weight of mice was equivalent to a typical dose for humans after conversion based on body surface area. In a separate study, treatment with JPYFII formulation was performed in the first 4 days during CS exposure and stopped before IAV infection to determine whether pre-infection treatment alleviated inflammation in the lungs.

Bronchoalveolar lavage fluid and lung collection
After euthanasia using sodium pentobarbital, the primary trachea and lungs of each mouse were cannulated and flushed once with 400 μl of phosphate-buffered saline (PBS) followed by three flushes with 300 μl of PBS. Total and differential leukocyte counts were obtained from the resulting bronchoalveolar lavage fluid (BALF). Cell types were identified based on standard morphological criteria. Blood was removed from whole lungs by perfusion of 5 ml of PBS through the right ventricle of the heart. Then they were rapidly excised en bloc, rinsed in PBS, blotted, snap-frozen in liquid nitrogen, and stored at −80 • C prior to further analysis.

Lung function measurement
Invasive pulmonary function devices (pulmonary function testing systems, DSI, Holliston, U.S.A.) were used to measure lung function of the mice. Briefly, each mouse was weighed and anesthetized with an intraperitoneal injection of pentobarbital (240 mg.kg −1 ) (Merck, Kenilworth, U.S.A.). Fifteen minutes after the induction of anesthesia, the trachea was cannulated using an 18-G, 30-mm intravenous plastic catheter, and a suture was used to seal the wall of the trachea around the cannula. The mouse was placed into a plastic chamber that was connected to a small-animal ventilator. After the breathing of the mouse was stabilized, forced expired volume over 0.1 s (FEV0.1), forced vital capacity (FVC), and functional residual capacity (FRC) of the lungs were measured in accordance with the manufacturer's instruction.

Histology
The left lung lobe from each mouse was removed without airway lavage and immersed in 4% paraformaldehyde for a minimum of 24 h. For the evaluation of mean linear intercepts (MLIs) evaluation, another eight mouse lungs in each treatment group were perfusion-fixed in situ for 10 min using 4% paraformaldehyde at 20 cm H 2 O pressure via a tracheal cannula and a constant-pressure syringe pump (Harvard, Holliston, U.S.A.). The lungs were left in situ for 1 h, then removed and immersed in 4% paraformaldehyde for a minimum of 24 h. After fixation, the lung tissues were processed in paraffin wax. Four-μm thick sections were cut with a microtome, mounted on glass microscope slides, processed to remove the paraffin, and stained with hematoxylin and eosin. The stained sections were viewed on a Nikon TI2-E microscope and photographed to perform tissue-scale imaging, followed by integrating the images with Nis-Elements software. The criteria used to generate inflammation scores were based on Curtis et al [32]. The composite image of each lung section was divided into nine parts, and the final score of each lung was the average score of the nine parts. MLI was also measured in nine random fields of view of each slide.

RNA extraction and quantitative real-time PCR
Total RNA was isolated from whole lung tissues using NucleoZOL reagent (MNG, Duren, Germany). RNA (1 μg) was reverse-transcribed into complementary DNA using HiScript ® Q RT SuperMix with gDNA wiper (Vazyme, China) in according with the manufacturer's instructions. Real-time PCR analysis was conducted using the Applied Biosystem (VI A7) with ChamQ SYBR ® Color qPCR Master Mix (Vazyme, China). β-Actin mRNA was used as an internal control. The threshold cycle (Ct) value is the PCR cycle number (out of 40) at which the measured fluorescent signal exceeds a calculated background threshold. It identifies the amplification of the target sequence value. The Ct value is proportional to the number of input target copies present in the sample. Ct numbers were transformed using Ct (threshold cycle time) and the relative value method and were expressed relative to β-actin mRNA levels.

Superoxide detection
Inflammatory cells in BALF were incubated with the chemiluminescent probe L-O12 (100 mM) (Wako Chemicals, U.S.A.) in the absence (for basal measurements) or presence of the PKC and NADPH oxidase activator phorbol 12, 13 dibutyrate (PDB, 1 mM) (Sigma Aldrich). Cells were dispensed into 96-well opti-plates. Luminescence was measured with a Multiscan Spectrum (Tecan Infinite M1000pro). Photon emission was recorded for 1 s for 30 cycles. Individual datapoints for each group were averaged from two replicates, subtracted against the average background values, normalized to total cell numbers, and expressed as relative light units.

Western blot analysis
A lysis buffer containing protease inhibitors was used to extract total protein from the frozen tissues. Protein concentrations were determined using a BCA assay kit. Proteins were separated by SDS-PAGE and transferred onto polyyinylidene difluoride membranes. The membranes were probed overnight with antibodies against β-actin (1:2,000 dilution, Cell Signaling Technology), heme oxygenase-1 (HO-1) (1:2,000 dilution, Abcam), and gp91 [phox] (1:2,000 dilution, BD Pharmingen). The membranes were then incubated with goat anti-rabbit or goat anti-mouse horseradish peroxidase-conjugated IgG at room temperature for 1.5 h. Immunoblots were detected and assessed using the Gel Doc XR+ Gel documentation system (Bio-rad). β-Actin was used as the housekeeping protein.

Viral titration
The whole lungs were removed aseptically, placed into bottles containing 1.0 ml of PBS with penicillin (1000 U.ml −1 ) and streptomycin (1000 μg.ml −1 ). Virus was extracted by grinding lungs followed by centrifugation. Plaques assays were performed as previously described [33]. Briefly, samples were diluted and incubated for 1 h on a Madin-Darby canine kidney (MDCK) cell monolayer to allow viral adsorption. Next, 1% agarose dissolved in MEM containing 1 μg/ml TPCK-treated trypsin was added to cells and incubated at 37 • C for 72 h. The cells were then fixed with 4% formaldehyde in saline for 1 h and stained with crystal violet. The plates were rinsed in water and plaques were counted. Viral titer was expressed as the PFU per lung.

Cytopathic effect assay of anti-IAV activity in vitro
Six hundred milligrams of JPYFII formulation was dissolved in 10 ml of MEM to prepare a 60 mg/ml stock solution, which was shaken overnight at 37 • C then filtered through a 0.22-μm syringe filter and stored at −20 • C for further use. Carboxylate oseltamivir (MW 312.4), which served as a positive agent, was purchased from Shanghai Roche Pharmaceuticals, Ltd. and dissolved in MEM at a concentration of 1 mg/ml. A/GZ/GIRD07/09 (H1N1) and B/Lee/1940 (FluB) were isolated from patient specimens at the Guangzhou Institute of Respiratory Health, China; A/HK/Y280/97 (H9N2), A/PR/8/34 (H1N1), and A/Aichi/2/1968 (H3N2) were kindly provided by Professor Zifeng Yang (Guangzhou Institute of Respiratory Health, China). All viruses were passaged in MDCK cells or 9-to 11-day-old specific pathogen-free chick embryos (purchased from Guangdong Wenshi Dahuanong Biotechnological Co., Ltd.) in a biosafety level 2 laboratory. MDCK cells were obtained from the Cell Bank of the typical Culture Preservation Committee of the Chinese Academy of Sciences (Shanghai, China), and routinely passaged in MEM supplemented with 10% fetal bovine serum, at 37 • C, and 5% CO 2 .
To evaluate the antiviral activities of JPYFII formulation against IAV, the cytopathic effect assay was used to examine the protective effect of JPYFII on cell morphological changes induced by different IAV strains in vitro. The half-maximal inhibitory concentration (IC 50 ) and selection index were utilized. A 90% MDCK monolayer was formed in 96-well plates, and the wells were divided into four groups: a normal group (MEM + MEM), a virus group (IVA + MEM), a positive control group (IVA + MEM with carboxylate oseltamivir), and a JPYFII treatment group (IVA + MEM with JPYFII). The cells were inoculated with 100-fold 50% tissue culture infective dose of the IAV dilution for 2 h at 37 • C. After adsorption, the supernatant was discarded, and the cells were washed twice with PBS. One hundred microliters of JPYFII formulation at the indicated concentrations (starting with 2000 μg/ml, >TC 50, followed by multiple dilutions until a concentration of 31.25 μg/ml was achieved) was added to the infected cells with 1.5 μg/ml N-Tosyl-L-Phenylalanine Chloromethyl Ketone (TPCK) treated-trypsin. After a 48-h incubation at 37 • C cell lesions were observed with an inverted microscope (Leica, Wetzlar, Germany), and the IC 50 of the JPYFII formulations were calculated according to the Reed-Muench method.

Statistical analysis
Grouped data are presented and expressed as mean + − standard error of the mean (S.E.M.). n represents the number of mice per treatment group. Statistical significance was determined by one-way or two-way analysis of variance (ANOVA) followed by Dunnett test or Sidak post hoc test for multiple comparisons. All statistical analyses were performed using GraphPad Prism for Windows (version 7.0). P<0.05 was considered statistically significant.

JPYFII formulation did not affect body weight in IAV-infected mice exposed to CS
The body weight of each mouse was measured and recorded daily to determine the effects of JPYFII formulation on body weight. Compared with sham + MEM + saline mice, the average body weight of CS + IAV + saline mice was slightly but significantly decreased, by 3.2% (P<0.05). JPYFII formulation did not change the body weights of sham + MEM or CS + IAV mice (P>0.05) ( Figure 1).

JPYFII formulation reduced the number of inflammatory cells in BALF from IAV-infected mice exposed to CS
Cellular inflammatory responses in the airways were examined by analyzing BALF from each mouse treated with either vehicle or JPYFII formulation (Figure 2A-E). The numbers of total cells, macrophages, neutrophils, and lymphocytes were significantly increased in BALF from CS + IAV + saline mice compared with sham + MEM + saline mice (P<0.05). JPYFII formulation reduced the numbers of total cells, macrophages, neutrophils, and lymphocytes in BALF from CS + IAV mice. JPYFII formulation did not affect the numbers of total cells, macrophages, neutrophils, or lymphocytes in BALF from sham + MEM mice.

JPYFII formulation alleviated airway inflammation but did not affect MLI or lung function in IAV-infected mice exposed to CS
Representative images in Figure 3A-L show hematoxylin and eosin-stained lung sections from mice treated with either vehicle or JPYFII formulation. Panels (A-D) depict sections of whole lung at the level of the porta pulmonis of the left lung lobes. Panels (E-H) depict inflammatory infiltrates in alveolar, peribronchial, and perivascular areas. Panels (I-L) depict the integrity and area of the pulmonary alveoli. CS exposure combined with IAV infection ( Figure 3C) resulted in severe pulmonary consolidation and inflammatory reactions composed of perivascular, peribronchial, and alveolar infiltrates of monocytes/macrophages, neutrophils, and lymphocytes followed by prominent inflammatory cell cuffing around bronchi and in the lung parenchyma ( Figure 3G).
Inflammatory scores in perivascular, peribronchial, and alveolar areas in CS + IAV + saline mice were significantly higher than those in sham + MEM + saline mice (P<0.01). JPYFII formulation reduced the area of pulmonary consolidation and inflammatory scores in alveolar regions (P<0.01) and peribronchial regions (P<0.05) in CS + IAV mice ( Figure 3D,H,M,N). JPYFII formulation did not reduce inflammatory scores in perivascular areas in CS + IAV mice ( Figure 3O). MLI, an indicator of alveolar size, was not increased in CS + MEM + saline mice compared to sham + MEM + saline mice. JPYFII formulation did not reduce the MLI in sham + MEM or CS + IAV mice ( Figure 3P). Representative pictures are shown in Figure 3I-L.
To evaluate the effects of JPYFII formulation on lung function in sham + MEM + saline mice and CS + IAV + saline mice, the FRC of lungs and the ratio of FEV0.1/FVC were measured using the animal pulmonary function testing system. FRC and FEV0.1/FVC ratio did not change significantly in CS + IAV + saline mice compared with sham + MEM + saline mice. JPYFII formulation treatment did not reduce FRC or increase the FEV0.1/FVC ratio in CS + IAV mice (Supplementary Figure S3).

JPYFII formulation alleviated superoxide production in BALF cells
Excessive oxidative stress is one of the main features of the lungs in patients with COPD and AECOPD [18]. To determine whether JPYFII formulation alleviated oxidative stress in the lungs of CS + IAV + saline mice, superoxide production was measured in BALF cells, and HO-1 and gp91 levels were measured in lung tissue. The production of superoxide from BALF cells was assessed ex vivo under PDB-stimulated conditions. Superoxide was significantly increased in CS + IAV + saline mice compared with sham + MEM + saline mice (7,218.68 + − 2,512.57 and 49.96 + − 15.18 respectively; P<0.01). In contrast, JPYFII formulation blocked excessive production of superoxide in CS + IAV + saline mice (169.12 + − 141.15; P<0.01) ( Figure 4A). Compared with sham + MEM + saline mice, protein levels of gp91 and HO-1 in mice exposed to CS + IAV were significantly increased, by 2.31-fold and 2.13-fold respectively (P<0.05). Administration of JPYFII formulation decreased the expression of both gp91 (1.35 + − 0.20; P<0.01; Figure  4B Superoxide production was assessed in BALF cells from sham + IAV + saline mice, sham + IAV + JPYFII mice, CS + MEM + saline mice, and CS + MEM + JPYFII mice. Superoxide was significantly increased in sham + IAV + saline mice compared with sham + MEM + saline mice (P<0.01), and superoxide levels were reduced by 57% (P>0.05) after JPYFII formulation treatment. CS exposure did not increase superoxide levels in the lungs of sham + MEM mice (Supplementary Figure S4).
Analysis of antiviral effects of JPYFII formulation using a cytopathic effect assay (Table 3)    Inhibition activity of JPYFII formulation on different strains of H1N1, H3N2, H9N2, and FluB was detected by CPE assay. IC50, TC50, and SI are described to evaluate the toxity and inhibitory activity of JPYFII formulation. SI is equal to the ratio of TC50 and IC50. SI > 2 indicates that the tested medicine is highly effective and low toxic. 1 < SI < 2 indicates that the tested medicine has low effectiveness and is highly toxic. SI<1 indicates the medicine is not effective. Data are shown as the mean + − S.E.M. from triplicate tests.

Pre-infection treatment with JPYFII formulation did not alleviate inflammation in lung
We have previously reported that JPYFII formulation alleviated inflammation and oxidative stress in CS-exposed mice [34], suggesting that it could provide protective effects against CS exposure. Given that CS worsened influenza virus (PR8 strain)-induced lung inflammation in a previous study [16], we hypothesized that JPYFII formulation treatment during CS exposure may reduce inflammation in CS + IAV-infected mice by protecting the epithelium from CS exposure. Pre

Discussion
Airway inflammation is one of the characteristic pathological changes observed in CS and infection-induced diseases (i.e., AECOPD). Both CS and viral infections lead to lung inflammatory responses, which include recruitment of immune cells into the airway, the release of proinflammatory cytokines, and an oxidant-antioxidant imbalance [35][36][37]. Infection with respiratory pathogens such as rhinovirus, influenza virus, and coronavirus reportedly aggravates inflammation in CS-exposed lungs [38,39]. Thus, resolution of inflammation is essential for the successful treatment of respiratory virus-associated AECOPD. In the present study, we investigated mice exposed to CS and infected with IAV to mimic the airway inflammation that occurs in AECOPD. JPYFII formulation reduced inflammatory cell influx and infiltration in the lungs, as well as expression of proinflammatory cytokines, chemokines, and oxidative stress in BALF cells and lungs in CS + IAV mice. It also reduced viral amounts in vivo and in vitro. Notably however, JPYFII formulation treatment prior to IAV infection did not reduce airway leukocyte influx or the release of proinflammatory cytokines in the lungs of CS + IAV mice.
In patients with AECOPD and animals challenged with IAV and CS, the inflammatory cell population in airways is increased [40,41]. Neutrophilia and lymphocythemia are indicators of an acute inflammatory response, especially in virus-associated conditions [42]. CS induces increased neutrophils and macrophages [43]. The leukocyte recruitments described above are usually driven by chemokines such as Cxcl1 and Cxcl2 among others, which are released by injured pulmonary epithelium [44]. In the present study JYPFII-treated CS + IAV mice exhibited fewer leukocytes, macrophages, neutrophils, and lymphocytes, indicating that JPYFII formulation alleviated inflammatory responses in the airways. This result was supported by histology observations indicating that JPYFII formulation reduced the infiltration of inflammatory cells, especially neutrophils and lymphocytes, in peribronchial and alveolar areas of the lungs in CS + IAV mice. The area of pulmonary consolidation was reduced in the CS + IAV mice that received JPYFII formulation treatment. These results indicated that JPYFII formulation facilitated the resolution of inflammation in the IAV-induced and CS-exposed lungs.
The mechanisms by which JPYFII formulation suppressed inflammatory cell influx in the lungs were investigated. The influx of immune cells into lungs depends on regulation by chemoattractants. Cxcl1 and Cxcl5, which are mainly expressed in epithelium, are neutrophilic chemokines [45] and were significantly increased in the CS + IAV mice. Ccl2, Cxcl2, and Cxcl10, the chemokines which recruit lymphocytes and monocytes, were also increased in CS + IAV mouse lungs. Therefore, increased expression of the chemokines in the lung tissue may explain the observed increases in macrophages, lymphocytes, and neutrophils that followed CS exposure and influenza infection. JPYFII formulation significantly reduced expression of Cxcl1, Cxcl2, Cxcl5, Cxcl10, and Ccl2 in CS + IAV mouse lungs but not sham + MEM mouse lungs. Thus, the JPYFII formulation may have reduced the number of leukocytes induced by CS combined with IAV by down-regulating the expression of chemoattractants.
Expression of chemokines was regulated transcriptionally via inflammatory pathways [46,47]. Blocking toll-like receptor (TLR) 2 or inhibiting MAPK and JAK/STAT pathway has been shown to suppress expression of Cxcl1, Cxcl5, and Cxcl8 [48]. Studies have shown that the ATF6 signaling pathway, which is activated by endoplasmic reticulum stress, is involved in the regulation of Cxcl1 [49]. In a previous study JPYFII formulation suppressed reticulum stress in cigarette smoke extract (CSE)-treated BEAS-2B cells [34], indicating that JPYFII formulation may down-regulate the expression of chemokines, for example Cxcl1, by dampening reticulum stress in CS + IAV mice.
Il-6 and Tnf-α are important inflammatory mediators that are involved in acute inflammatory flare ups [50]. Tnf-α, along with granulocyte-marophage colony-stimulating factor can activate neutrophils and regulate neutrophil survival via the NFκB and JAK/STAT signaling pathway [51]. In the present study Il-6 and Tnf-α mRNA expression were reduced in CS + IAV mice after JPYFII formulation treatment, demonstrating that the acute inflammatory response was suppressed by JPYFII formulation treatment. The underlying mechanism may be associated with inhibition of NFκB signaling pathway, which has been proved to be a target of JPYFII formulation in a CS+LPS mouse model [24]. Proinflammatory mediators were also detected in sham + IAV mice and CS + MEM mice with or without JPYFII formulation treatment. JPYFII formulation decreased Il-6, Tnf-a, and Il-1β expression in sham+IAV mice, but did not affect expression of pro-inlfammatory mediators in CS + MEM mice. These results demonstrated that the production of inflammatory mediators induced by IAV might be an important target in JPYFII formulation treatment.
Hypermetabolism induced by inflammation is one of the reasons for body-weight loss [52]. However, the body weight loss was not reduced in JPYFII treated CS + IAV mice compared with sham + MEM mice. As reported, nicotine in CS has a great effect on appetite suppression [53]. Thus, the data demonstrated that JPYFII formulation suppressed the inflammation in lungs of CS+IAV mice but did not alleviate the suppression of appetite by nicotine.
Excessive production of superoxide is a feature of AECOPD. To determine whether JPYFII formulation reduced oxidative stress in the lungs and investigate the mechanism involved, superoxide production was measured in BALF cells. gp-91 protein is a subunit of NADPH oxidase 2. Heme oxygenase-1 (HO-1), which can protect epithelium and endothelium from excessive oxidative stress induced by CS [54,55], and gp91 protein expression was also measured in the present study. Superoxide production, and the expression of gp-91were significantly reduced by JPYFII formulation in CS + IAV-infected mice. NADPH oxidase 2 is a main superoxide generating enzyme expressed in lungs. gp91-deficient mice reportedly exhibit reduced production of superoxide radicals and decreased levels of lung inflammation after IAV infections [56,57]. Thus JPYFII formulation-induced reduction of superoxide and inflammation in lungs may be related to its effect on reduction of Nox2 expression. HO-1 is regulated by NRF2 and the transcription regulator protein BACH1 [58]. Down-regulation but not up-regulation of HO-1 in the lungs of CS + IAV mice treated with JPYFII formulation supported the speculation that JPYFII formulation suppressed superoxide production instead of protecting lung tissue form endogenic superoxide via the NRF2 pathway. Viral infection leads to oxidative stress in lung tissue, and anti-oxidative stress is an important target that can be used to improve the viral clearance [57,59]. JPYFII formulation could reduce superoxide production in both sham + IAV mice (P>0.05, Supplementary Figure S4) and CS + IAV mice ( Figure 4A). As well as JPYFII formulation reducing inflammatory cells and mediators, it is possible that JPYFII clears influenza virus in the lungs. In the present study JPYFII formulation decreased viral load and expression of IAV nucleoprotein in CS + IAV mice. In in vitro experiments JPYFII formulation reduced H1N1, H3N2, and H9N2 viral loads.
Reduction of H 2 O 2 production weakens suppression of the expression of type I IFNs mediated by the TLR7/TLR3 signaling pathway [60,61]. Therefore, we speculated that JPYFII formulation may increase the expression of IFNs. However, JPYFII formulation did not increase the expression of Ifnα or Ifnβ in CS + IAV mice or sham + MEM mice. On the contrary, JPYFII formulation reduced the expression of IFNs (Ifnα, Ifnβ, Ifnγ, and Ifnλ) in sham + IAV mice and CS + IAV mice. These results indicated that clearance of IAV by JPYFII formulation may not be dependent on IFN-related mechanism.
IFNs can contribute to a general antiviral response via the activation of a broad range of effector molecules [62]. DAT is a small-molecule compound that can augment type I IFN signaling. In a recent study treatment with DAT prior to IAV infection yielded better protection against death than DAT treatment post-IAV infection [63]. Notably, human interferon inducible transmembrane protein 3 inhibited IAV-induced inflammation [64]. These studies demonstrated that drugs that increase the expression of IFNs can prevent viral infections and inflammation induced by them. In the present study JPYFII treatment prior to IAV infection did not reduce inflammation in the lungs of CS + IAV mice, which is circumstantial evidence that clearance of IAV by JPYFII formulation is not dependent on augmentation of IFNs.
Drugs that inhibit viral replication in host cells can decrease viral loads after viral infection. Quercetin 3'-glucuronide (Q3G) is a derivative of quercetin in Radix astragali, and it reportedly exhibits strong binding affinity to the cap-binding site of the PB2 subunit responsible for viral replication [65]. Thus, inhibition of IAV replication may be the underlying mechanism by which JPYFII reduces the amount of IAV in CS-exposed mice.
One limitation of the present study is that we could not determine whether JPYFII formulation could improve lung function or protect structure of alveoli in CS + IAV mice. This is because 4 days of smoke exposure is too short to induce airway narrowing or destruction of the airways. Four days of smoke exposure did induced abnormal immune responses however, similar to that which occurs in COPD lungs, as has been reported in a previous study [30,66]. The effects of CS can persist after smoking cessation [16]. In the present study 4 days of CS exposure suppressed the expression of IFNs (Type I, TypeII and Type III) by 71-84% of the amount observed in sham mice 5 days after smoke exposure was stopped, which was because CS reduced IFN levels by disrupting TLR3 cleavage [67,68]. In previous studies prior CS exposure has exacerbated lung inflammation in influenza virus-infected mice [59]. In the present study, there was more expression of chemokines and Mmp12 in CS + IAV mice than in sham + IAV mice, which is concordant with a previous study [59]. We used a mouse model of CS exposure combined with IAV to evaluate the anti-inflammatory and anti-viral effects of JPYFII formulation. By using this experimental model, we revealed the potential benefit of JPYFII formulation on CS and IAV-induced inflammatory disease, i.e., AECOPD.
The reason why CS aggravates influenza virus-induced inflammation is still unclear. CS exposure reportedly leads to increased neutrophilic inflammation in the airways, and delayed viral clearance due to a reduced respiratory syncytial virus-specific CD8+ T cell response [69]. In another study mice exposed to CS recovered poorly from primary Influenza A pneumonia with reduced type I and II IFNs and virus-specific immunoglobulins [70]. Thus, the underlying mechanism may be related to down-regulation of immune responses to viruses induced by CS. In the present study, CS exposure did not induce the expression of acute inflammatory cytokines because CS exposure had been stopped for 5 days by the time the lung tissue was taken. Even so, expression of chemokines and IFNs was reduced in CS + MEM mice (P>0.05; Table 2), which indicates persistant suppression of immune system by CS. Pre-treatment with JPYFII formulation did not relieve IAV-induced airway inflammation in CS + IAV mice, indicating that JPY-FII formulation may not improve the immunosuppression induced by CS. This speculation was supported by the oberservation that JPYFII did not increase expression of chemokines or IFNs in CS + MEM mice ( Table 2).

Conclusion
In the present study, JPYFII formulation reduced airway inflammation in IAV-infected CS-exposed mice. Thus, the study suggests that JPYFII formulation may be a potential drug for the effective treatment of AECOPD associated with IAV infection.