Role of PM2.5 in the development and progression of COPD and its mechanisms

Background A multitude of epidemiological studies have shown that ambient fine particulate matter 2.5 (diameter < 2.5um; PM2.5) was associated with increased morbidity and mortality of chronic obstructive pulmonary disease (COPD). However, the underlying associated mechanisms have not yet been elucidated. We conducted this study to investigate the role of PM2.5 in the development of COPD and associated mechanisms. Methods We firstly conducted a cross-sectional study in Chinese han population to observe PM2.5 effects on COPD morbidity. Then, in vitro, we incubated human bronchial epithelial cells to different concentrations of PM2.5 for 24 h. The expression levels of IL-6 and IL-8 were detected by ELISA and the levels of MMPs, TGF-β1, fibronectin and collagen was determined by immunoblotting. In vivo, we subjected C57BL/6 mice to chronic prolonged exposure to PM2.5 for 48 weeks to study the influence of PM2.5 exposure on lung function, pulmonary structure and inflammation. Results We found that the effect of PM2.5 on COPD morbidity was associated with its levels and that PM2.5 and cigarette smoke could have a synergistic impact on COPD development and progression. Both vitro and vivo studies demonstrated that PM2.5 exposure could induce pulmonary inflammation, decrease lung function, and cause emphysematous changes. Furthermore, PM2.5 could markedly aggravated cigarette smoke-induced changes. Conclusions In short, we found that prolonged chronic exposure to PM2.5 resulted in decreased lung function, emphysematous lesions and airway inflammation. Most importantly, long-term PM2.5 exposure exacerbateed cigarette smoke-induced changes in COPD.


Background
Chronic obstructive pulmonary disease, a major public health problem worldwide, is characterized by irreversible and progressive airflow limitation that is associated with chronic and aberrant pulmonary inflammation and pulmonary remodeling induced by abnormal pulmonary response to inhaled noxious particles and gases [1]. In recent years, COPD has been a major cause of global morbidity and mortality. It will become the third leading cause of death and the fifth global burden worldwide by 2020 [2]. Cigarette smoke has been widely recognized as the most important causal factor for COPD for decades, yet only a small part of smokers (15-20%) could eventually develop COPD [3,4]. Moreover, approximately 25% of COPD patients are non-smokers [5]. Thus, it suggests that other factors also contribute significantly to COPD development and progression. In recent decades, substantial epidemiological evidence indicated that ambient particulate air pollution including PM 2.5 is a major detrimental risk factor for COPD [6][7][8][9]. Moreover, numerous epidemiological studies from large population cohorts demonstrated that PM 2.5 was not only associated with an enhanced risk of COPD hospitalization, morbidity and mortality [10][11][12], but also exacerbated and aggravated respiratory function and symptoms of COPD patients, such as shortness of breath, coughing and wheezing [13][14][15]. Air pollution improvement would substantially provide benefits to COPD patients [16][17][18]. However, little is known about the mechanisms underlying these effects that PM 2.5 induces and further aggravates COPD development and progression. Therefore, we conducted this study to explore the influence of chronic prolonged PM 2.5 exposure on COPD development. Firstly, we conducted a cross-sectional study in Chinese han population to observe PM 2.5 effects on COPD morbidity and on smokers with COPD. Then, in vitro, we incubated HBEs with different concentrations of PM 2.5 to investigate the role of PM 2.5 in COPD development and its mechanisms. Additionally, to further deeply explore the impacts and associated mechanisms of PM 2.5 on COPD, we conducted chronic prolonged PM 2.5 -exposed animal model for 48 weeks to mimic COPD development over a substantial duration of individual life so as to study the influence of PM 2.5 exposure on lung function, pulmonary structure and inflammation to further acquire in-depth appreciation of the associated mechanisms of PM 2.5 on COPD.

Subject
Study subjects were recruited from urban (Wuhan) and rural (Haokou, Qianjiang) settings. The urban cohort including 1372 participants recruited from two communities in Wuhan. The rural cohort including 1843 participants recruited from 15 villages in Haokou, about 200 km away from Wuhan. Study subjects were born in the local area and have resided locally for at least five years. Individuals were diagnosed as COPD on the basis of American Thoracic Society criteria and the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria. The inclusion and exclusion criteria of COPD subjects in this study were described in our previous study [19,20]. Briefly, postbronchodilator forced expiratory volume in one second (FEV1)/forced vital capacity (FVC) < 70%, and FEV1 < 80% predicted was the diagnostic criteria in this study. Subject were excluded from the study if they were diagnosed as asthma and other obstructive syndromes. The existing PM 2.5 levels per year at the two study sites were obtained from Hubei Provincial Environmental Monitoring Center station (HEWC), which was reported in detail in our previous study [20]. The data were gathered automatically and continuously, 24 h/d and 365 d/y, with no interruption. The monitoring was carried out by the HEMC and authorized by the United States Environmental Protection Agency (EPA).
Written informed consents were obtained from all participants, and the study was approved by the institutional ethics committees of local participating hospitals. PM 2.5 sampling collection and extraction PM 2.5 was continuously collected using UAS-310-PM 10-2.5 and PM 2.5 high-volume air sampler (MSP, USA) at a flow rate of 300 L /min from ambient air in the campus of Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China on Teflon microporous membranes and stored at − 20°C in the dark until extracted. The filters containing PM 2.5 were divided into small pieces of approximately 1 cm 2 and then immersed in sterile purified water before sonicating for 4 × 30 min in a water bath with an ultrasonic cleaner. After drying by lyophilisation, PM 2.5 dry powder was stored at − 80°C until use.

Cell culture and stimulation
HBEs were purchased from ATCC (Manassas, VA, USA) . Cells were cultured in 1640 RPMI medium (HyClone, USA) containing 100 units/mL penicillin and 100 μg/mL streptomycin (KeyGEN, Nanjing, China) and supplemented with 10% fetal bovine serum (Gibco, USA). The cells were maintained in a humidified atmosphere of 5% CO 2 at 37°C. Before stimulation, cells were cultured in serum-free medium for 24 h to arrest and synchronize the cell growth. Then they were treated with different final concentrations of freshly dispersed PM 2.5 preparations with or without cigarette smoke extract (CSE, 25 μg/ml) (Murty Pharmaceuticals, Inc., Lexington, KY, USA) for 24 h. After 24 h stimulation, cell culture supernatants were collected, centrifuged and stored at − 80°C for futher analysis.

Western blot analysis
Total proteins were collected in appropriate ice-cold RIPA lysis buffer for immunoblotting analysis. Anti-MMP9 Ab (Santa Cruz, CA), anti-MMP12 Ab (Abcam, Cambridge, UK), collagen I (Abcam, Cambridge, UK), collagen III (Abcam, Cambridge, UK), anti-TGFβ1 Ab (Proteintech Group, China), fibronectin (Proteintech Group, China) and anti-β-actin (Proteintech Group, China) were used. Image J was used to quantify the intensity of the protein band, which was normalized to βactin in the analyses.

Animals
Six-week-old male C57BL/6 mice were obtained from Hubei Provincial Laboratory Animal Public Service Center, Wuhan, Hubei Province, China. They were equilibrated for 2 weeks before exposure experiment. The mice were fed a normal diet and housed in cages at the Experimental Animal Centre, Tongji Medical College, Huazhong University of Science and Technology. All animal experimental procedures were authorized by Huazhong University Animal Experiment Ethics Committee and executed in accordance with institutional regulations for ethical animal use.
Animal experimental procedures C57BL/6 mice were randomly divided into four groups (n = 6 per group): control group, CS-exposed group, PM 2.5 -exposed group (PM 2.5 ) and PM 2.5 + CS exposed group (PM 2.5 + CS). The CS-exposed procedure was carried out as described previously [21]. Briefly, CSexposed mice were exposed to cigarette smoke produced by 5 cigarettes twice daily, 5 d/wk., using a PAB-S200 Animal Passive Smoking Whole Body Exposure System (BioLab Technology Co. Ltd., Beijing, China) for continuously for 48 weeks. PM 2.5 -exposed mice were placed in a 192 L exposure chamber (0.8 m in length, 0.6 m in width and 0.4 m in height) with three holes of 10 mm in diameter on the top panel of the chamber for fresh air inlet and one hole of 30 mm in diameter on the side panel of the chamber, through which PM 2.5 aerosols generated by an ultrasonic nebulizer to the exposure chamber. PM 2.5 -exposed C57BL/6 mice, in an awake and unrestrained state, were subjected to concentrated ambient air PM 2.5 in whole-body inhalation protocol in vivo for continuously for 48 weeks. The PM 2.5-exposed procedures were referred as described previously [22]. In detail, the ambient daily average PM 2.5 concentration close to our study site was 101.3μg/m 3 , which was much higher than that listed in the Global Air Quality Guidelines from the World Health Organization (annual average: 10 μg/m 3 ). The average PM 2.5 concentration in the exposure room was 560μg/m 3 , equivalent to 5.7-fold level of ambient daily PM 2.5 . Since the mice were exposed for 6 h/d for 5 d/wk in concentrated PM 2.5 exposure group, the equivalent PM 2.5 concentration for mice exposed in the exposure room, after normalization over the exposure study period, was 110 μg/m 3 , which was within the annual average concentrations PM 2.5 . At the same time, mice in PM 2.5 + CS exposed group was exposed to cigarette smoke followed by PM 2.5 as the methods described above. Simultaneously, the controls were exposed to filtered air.

Lung function measurement
Lung function was measured by using AniRes 2005 lung function system (Bestlab, Beijing, China) as previously described [23]. In detail, the mice were anaesthetized with 1% pentobarbital sodium by means of intraperito-neaI injection and subsequently a tracheal cannula was inserted. Then, the mice were immediately placed supine in a sealed whole-body plethysmograph and connected to a computer-controlled ventilator through the tracheal cannula. All animals were ventilated mechanically at respiratory rates of 90 breaths/min with a tidal volume of 5 ml/kg and at expiration/inspiration time ratio of 1.5: 1.0 with a computer-controlled and assisted small animal ventilator. Forced expiratory vital capacity (FVC), Forced expiratory volume in 0.1 s (FEV 0.1 ) and FEV 0.1 / FVC were measured after the mouse was acclimatized for at least 30s and all parameters were stable.

Pulmonary morphometric and pathology assessment
After pulmonary function measurements, the chests of the mice were immediately opened, and the tracheas were intubated with a tracheal cannula. Before performing bronchoalveolar lavage (BAL), we ligated the left main bronchus to avoid any impacts of operations on pulmonary morphometric and pathology assessment. Then, the right lung was instilled with 0.8 ml of PBS X 3 via PE-90 tubing to acquire the BAL fluid. The first BAL fluid was centrifuged to remove cells, and the cell-free supernatants were collected and stored at − 80°C for ELISA analysis. Afterwards, the right lung was excised and stored at − 80°C for further analysis. The left lung was not lavaged, but in contrast was inflated and fixated with 4% paraformaldehyde through the tracheal cannula, immersed in fresh 4% paraformaldehyde after excision, and then embedded in paraffin. Lung tissue was stained with hematoxylin and eosin (H&E) and analyzed by light microscopy for histology and morphometric examination. We determined enlargement of alveolar spaces by measuring mean linear intercept (MLI) as described previously [24]. Additionally, collagen deposition was determined by using Sirius Red according to the manufacturer's instructions. Peribronchial collagen deposition in lung tissue of mice was assessed as previously described [25]. In detail, the length of the basement membrane (Pbm) was manually marked on the digital representation of the airway. For the quantification of collagen deposition, the software (KS400; Zeiss) was used to determine the area in the airway wall covered by collagen and its value was counted. The area of collagen deposition was normalized to Pbm. Three lung sections per animal in every experimental group were examined.

Immunohistochemistry
The lung sections acquired from formalin-fixed, paraffin-embedded left lung lobes were stained with anti-TGF-β1-Ab (Abcam, Cambridge, UK), anti-MMP9-Ab (Santa Cruz, CA), anti-MMP12-Ab (Proteintech, China) and anti-fibronectin Ab (Proteintech, China), respectively, according to the manufacturers' instructions. As described previously [23], the mean integral OD of MMP9, MMP12, TGF-β1 and fibronectin protein staining in the bronchiolar epithelium was detected. The mean integral OD of protein staining was the ratio of the integral OD of protein staining-positive epithelium to the area of corresponding bronchial epithelium. The same parameters was used for all of the images.

Enzyme linked immunosorbent assay (ELISA)
The cytokine and chemokine protein levels in the BALF and cell culture supernatants were determined using ELISA. Human IL-6 and IL-8 in cell culture supernatants and mouse IL-6 and KC (mouse IL-8) in BAL were separately determined by ELISA kits from R&D Systems Inc. (Minneapolis, MN, USA) according to the manufacturer's instructions.

Statistical analysis
All statistical analyses were carried out by using IBM SPSS 19.0 and GraphPad Prism5.0. Categorical variables were expressed as counts and percentages, and chisquare tests were used to compare these variables. Continuous variables were reported as mean ± standard deviation (SD). Normally distributed variables were assessed by using Student's t test for two groups or one-way ANOVA with Newman-Keuls post hoc test for multiple comparisons and abnormal distribution variables were evaluated by Mann-Whitney U-test. Multiple logistic regression analysis were used to find out potential confounders associated with COPD onset. P < 0.05 was considered as statistical significance.

Subject characteristics
The baseline characteristics of the cohorts are summarized in Table 1. Briefly, the urban cohort included 1372 participants, and the rural cohort included 1843 participants. Of these subjects, 73.8% were male in the urban cohort and 79.2% were male in the rural cohort. Among the urban participants, 86.1% were over 40 years old, and 87.7% were over 40 years old in the rural cohort. There were no statistically significant differences in age distribution between the two cohorts. The percentage of those who never-smoke in the urban cohort was markedly higher than that of the rural population (urban, 74.3% vs 40.5%, rural). Smokers in the urban cohort was significantly lower than that in the rural cohort (P < 0.05). Moreover, 18.9% of the urban cohort had a smoking history above 10 pack-years, which was significantly lower than the rural cohort. The subjects in the study were all born in their current city/area and lived there for more than five years.
The association between PM 2.5 and the incidence of COPD As shown in Fig. 1a, mean PM 2.5 concentration in the urban was obviously higher than the rural (101.3 μg/ml vs 59.4 μg/ml). Meanwhile, in non-smoking cohorts, COPD incidence in the urban was higher than that of the rural, while not statistically significant (Fig. 1b). Interestingly, compared to never-smoking cohorts, COPD incidence in cigarette smoking cohorts increased largely in both areas. Multiple logistic regression analysis were used to find out potential confounders (including age, gender, smoking status as well as PM 2.5 explore) associated with COPD onset. The results was that no statistical correlation of the other confounders and COPD, except smoking and PM 2.5 . Taken together, the results indicated that the effect of PM 2.5 on COPD incidence in the population could be associated with its atmospheric concentration. Moreover, PM 2.5 and cigarette smoking could have a synergistic effect on COPD development and progression.

PM 2.5 augmented inflammatory cytokines and growth factors expression in HBEs
Cytokines and chemokines play important roles in the development of COPD [26,27]. To examine the proinflammatory potential of PM 2.5 , we assessed the production of cytokines by measuring IL-6 and IL-8 levels in cell supernatants. HBEs were incubated with 25 μg/ml to 200 μg/ml PM 2.5 either with or without CSE for 24 h. Then, IL-6 and IL-8 levels in cell culture supernatants were determined by ELISA. With the lowest concentration of PM 2.5 (25μg/ml) examined, the levels of IL-6 was increased from a basal level of 258.08 ± 75.91 pg/ml up to 346.78 ± 84.43 pg/ml and the levels of IL-8 from a basal level of 290.36 ± 103.61 pg/ml up to 443.16 ± 104.10 pg/ml (Fig. 2a, b). Moreover, a concentration-response effect were observed in the release of IL-6 and IL-8 in response to PM 2.5 , reaching 613.31 ± 63.72 pg/ml for IL-6 and 1238.99 ± 200.81 pg/ml for IL-8, with the highest concentration of PM 2.5 (200 μg/ml). Moreover, compared to CSE-exposed group, PM 2.5 could significantly enhanced the levels of IL-6 production from 418.737 ± 42.74 pg/ml to 555.78 ± 61.89 pg/ml and the levels of IL-8 production from 398.61 ± 94.01 pg/ml to 488.70 ± 92.51 pg/ml at the lowest concentration of PM 2.5 (25μg/ml) co-stimulated with CSE ( Fig. 2c and d). A dose-response effect was also observed by co-stimulation of all four concentrations of PM 2.5 and CSE, with the level of IL-6 up to 673.68 ± 13.82 pg/ml and IL-8 up to 1260.625 ± 147.00 pg/ml by using the highest concentration of PM 2.5 (200 μg/ml) together with CSE. Similarly, PM 2.5 exposure could also obviously upregulate MMP9, MMP12, TGF-β1, fibronectin and collagen protein expression at a concentration-dependent manner and increased CS-induced these proteins expression (Fig. 3a, b). All the above results indicated that PM 2.5 could induce the enhanced effects of CS-induced inflammatory cytokine secretion and growth factors expression.

PM 2.5 exposure impaired pulmonary function in mice
Assessment of the lungs after PM 2.5 /CS exposure for 48 weeks indicated significant lung function decline. As shown in Table 2, ratio of FEV 0.1s to forced vital capacity (FEV 0.1s /FVC) in PM 2.5 / CS-exposed mice were all less than 70%, while the controls were greater than 70%. Besides obvious decrease of FEV 0.1s /FVC in PM 2.5 and CSexposed group individually, we observed that FEV 0.1s / FVC was significantly decline in PM 2.5 + CSE group, implying that PM 2.5 and CS exposure might synergistically lead to greater damage to pulmonary function than PM 2.5 /CS treatment alone. In addition, the FEV 0.1 of PM 2.5 + CS-exposed mice significantly decreased compared to that of any other group, although there were only slight decreases in the FEV 0.1 in PM 2.5 /CS-exposed mice, and no statistically significant difference was found in comparison with control mice. Interestingly, PM 2.5 / CS-exposed groups had slightly increased FVC values, while the PM 2.5 & CS-exposed mice had slightly decreased FVC values.

PM 2.5 exposure induced pulmonary emphysema changes: airspace enlargement
Compared to control group, PM 2.5 /CS exposure markedly led to airspace enlargement, respectively, as clearly demonstrated by histological examination of mice lung tissue sections stained with hematoxylin and eosin (HE) in Fig. 4. Remarkably, a clearly stronger enlargement in alveolar spaces was induced by PM 2.5 + CS exposure in comparison with the other three groups. To further Fig. 1 The association of PM 2.5 with COPD incidence in the population study. a The levels of PM 2.5 in urban and rural settings; b COPD incidence among different groups based on smoking-pack-years The results are mean ± SD of three independent experiments. *P < 0.05, ** P < 0.01, *** P < 0.001 vs control group; # P < 0.05, ## P < 0.01, ### P < 0.001 vs CSE group quantify the differences in airspace enlargement between groups, we also measured mean linear intercept (MLI), which has been recognized as a good and accurate indicator for measuring emphysematous changes [28,29]. An obvious increase in MLI was observed both in chronic PM 2.5 and CS exposure groups, and a markedly greater airspace enlargement was induced in PM 2.5 + CS exposure group, confirming emphysematous changes under chronic PM 2.5 with or without cigarette smoke exposure.

PM 2.5 exposure induced pulmonary inflammation and aggravated CS-induced inflammatory changes in mice
COPD is characterized by pulmonary emphysema and inflammation. To assess pulmonary inflammation, lung histopathology and BALF pro-inflammatory cytokine levels were detected. As shown in Fig. 5, it revealed that there was striking pulmonary inflammation in the peribronchial, perivascular and alveolar spaces of the lung in PM 2.5 /CS exposure. Moreover, PM 2.5 + CS exposure group showed slightly increased inflammation compared to PM 2.5 /CS exposure (Fig. 5a-d). In addition, the levels of pro-inflammatory cytokines IL-6 and KC (IL-8 functional homologs) in the BALF supernatant were clearly elevated separately in PM 2.5 -exposed (IL-6,CON 32.50 pg/ml vs PM 2.5 57.89 pg/ml, p = 0.0082)(KC,CON 49.65 pg/ml vs PM 2.5 72.01 pg/ml, p = 0.0305) and CS-exposed group (IL-6,CON 32.50 pg/ml vs CS 52.00 pg/ml, p = 0.0029)(KC,CON 49.65 pg/ml vs CS 80.70 pg/ml, p = 0.0248) compared to control group. Compared to PM 2.5 /  Values are mean ± SD, n =6 mice per group; * P < 0.05 in comparison to control group; # P < 0.05 in comparison to PM 2.5 + CS exposure group Fig. 4 Morphological and pathological changes of lung tissue in mouse model. The representative micrographs of H&E-stained lung tissue of long-term exposure to PM 2.5 with or without CS for 48 weeks (a-d) (200x). a control group; b CS-exposed group; c PM 2.5 -exposed group; d PM 2.5 + CS -exposed group. e Quantitative morphometric measurements of MLI. ** P < 0.01, *** P < 0.001 vs control group Fig. 5 Effect of PM 2.5 and cigarette smoke exposure on pulmonary inflammation. The representative changes of pulmonary inflammation after long-term exposure to PM 2.5 with or without CS for 48 weeks (a-d) (200x). a control group; b CS-exposed group; c PM 2.5 -exposed group; d PM 2.5 + CS -exposed group. The levels of inflammatory cytokines IL-6 and KC (mouse IL-8) in BAL were detected by ELISA; e the level of IL-8 in BAL and f the level of IL-6 in BAL. Data are presented as mean ± SD (n = 6). * P < 0.05, ** P < 0.01 vs control group CS-exposed group, PM 2.5 + CS exposure group showed slightly increased levels of IL-6 and KC, although there was no statistically significant difference. PM 2.5 enhanced MMP9, MMP12 and TGF-β1 protein expression in mice lung In present study, the effects of PM 2.5 /CS exposure in modulating the protein expression levels of MMP9, MMP12 and TGF-β1 were determined by immunehistochemical analysis and Western blots in mice lung sections (Figs. 6 and 7). Compared with control group, PM 2.5 /CS -exposed group induced significant increases in MMP-9 and MMP12 protein expression evaluation by immunohistochemistry (Fig. 6) and by immunoblot (Fig. 7). Moreover, the protein expression levels of MMP9 and MMP12 in PM 2.5 + CS group were significantly enhanced relative to any other stimulus group. In contrast, TGF-β1 protein levels in PM 2.5 /CS group were only slightly increased compared with control group. Although TGF-β1 protein expression was significantly increased in PM 2.5 + CS group compared with that of PM 2.5 group, there had no significant difference between PM 2.5 + CS group and PM 2.5 group.
Effect of PM 2.5 and cigarette smoke exposure on airway wall remodeling To further detect the impact of PM 2.5 exposure on pulmonary extracellular matrix deposition in the airway wall, fibronectin and collagen were evaluated by immunochemistry and Sirius Red, respectively. An evident enhancement in peri-bronchial fibronectin deposition was observed in PM 2.5 /CS group relative to controls (Fig. 8). Similarly, PM 2.5 /CS exposure markedly upregulated the expression of collagen in peri-bronchial areas. However, fibronectin and collagen protein expression levels in PM 2.5 + CS group had no clearly significant increase compared to CS group.

Discussion
COPD is a preventable and treatable pulmonary disease with increasing prevalence worldwide [1]. Numerous epidemiologic studies have revealed the associations between particulate pollution exposure including PM 2.5 and increased morbidity and mortality of cardiopulmonary diseases, in both developed as well as developing countries [30][31][32]. However, it remains poorly understood about the underlying effects and associated mechanisms by which PM 2.5 exposure induces and aggravates COPD development. In this study, we demonstrated that chronic prolonged PM 2.5 exposure impaired lung function, triggered emphysematous lesions and induced pulmonary inflammation and as well as airway wall remolding. Most importantly, prolonged PM 2.5 exposure can obviously worsen CS-induced changes of COPD, offering evidence for an important association between PM 2.5 and cigarette smoking in COPD development and progression. Fig. 6 Immunohistochemical staining of MMP9, MMP12, and TGF-β1 in lungs sections induced by PM 2.5 exposure (400x) (a) and semiquantitative assessment of MMP9, MMP12, and TGF-β1 protein expression using Image-Pro Plus (b). Data are presented as mean ± SD (n = 6). *P < 0.05, ** P < 0.01, *** P < 0.001 vs control group; # P < 0.05, ## P < 0.01, ### P < 0.001 vs PM 2.5 + CSE-exposed group A 10-μg/m 3 increase of ambient PM 2.5 concentration could be related to a 2.5% increase of COPD mortality and 3.1% increment of COPD hospitalizations [33]. However, the extent of relations among different geographic locations varies [31]. In this study, we found that in never-smokers, COPD incidence in the urban cohort was higher than that of the rural cohort. Intriguingly, relative to never-smokers, COPD incidence in the smoking subgroups in the two areas both increased greatly regardless of smoking pack-years. We discovered that PM 2.5 concentration in the urban was obviously higher than that in the rural. The results demonstrated that the role of PM 2.5 on COPD incidence in the population was associated with its ambient concentration and smokers were much more susceptible to subject to PM 2.5 pollution relative to non-smokers. In other words, PM 2.5 and cigarette smoke could have a synergistic effect on COPD development and progression.
To gain insight into the mechanisms associated between PM 2.5 and COPD incidence, HBEs were exposed to increased doses of PM 2.5 with or without CSE to evaluate the levels of pro-inflammatory mediators and important protein markers involved in airway remolding of COPD.
Consistent with the results of previous studies, our findings demonstrated that the levels of IL-6 and IL-8 were upregulated in PM 2.5 -exposed cells at a dose-dependent manner [34,35]. Similarly, PM 2.5 exposure could also markedly increase MMP9, MMP12, fibronectin, collagen and TGF-β1 protein expression at a concentrationdependent manner. In addition, PM 2.5 could further strengthen these effects induced by CSE. These results certified that PM 2.5 not only markedly enhanced the levels of pro-inflammatory cytokines, MMP9, MMP12, fibronectin, collagen and TGF-β1 protein expression, but also reinforced these effects induced by CSE.
To further deeply elucidate the impacts and associated mechanisms of PM 2.5 on COPD progression and development, other than studies by intra-tracheal or nasal exposure methods [36,37], we created a chronic prolonged PM 2.5 -exposed animal model for 48 weeks to better simulate the development of COPD over a substantial duration throughout individual life. We discovered that PM 2.5 /CS exposure could markedly induce FEV 0.1 /FVC decline, FEV 0.1 decrease and FVC increase, indicating that PM 2.5 /CS exposure might lead to lung function decline. Moreover, PM 2.5 + CS remarkably Fig. 7 Western blot examination for MMP9, MMP12, and TGF-β1 protein levels expression in lungs sections induced by PM 2.5 /CS exposure. The expressions of MMP9, MMP12, and TGF-β1 were analyzed with western blotting (a) and (b-d) the proteins expression were evaluated. Data are expressed as mean ± SD (n = 6). *P < 0.05, ** P < 0.01, *** P < 0.001 vs control group; # P < 0.05, ## P < 0.01, ### P < 0.001 vs PM 2.5 + CSE-exposed group decreased FEV 0.1 /FVC and FEV 0.1 relative to their onlycounterpart. These results indicated that chronic PM 2.5 exposure might impair lung function and have a synergistic role with cigarette smoke in triggering much more serious pulmonary function decline.
The persistent airflow limitation characteristic of COPD is caused by chronic inflammation, narrowing of peripheral airways and parenchymal destruction due to emphysema [4]. In the present study, PM 2.5 /CS markedly induced airspace enlargement, alveolar wall destruction and striking pulmonary inflammation. Importantly, we discovered that PM 2.5 could aggravate CS-related pulmonary emphysematous changes and inflammation.
An imbalance between proteinases and their inhibitors is believed to have a key role in alveolar destruction and airway remolding in COPD development [25,38]. In addition, TGF-β1 also has an important effect on lung remodeling [39][40][41]. PM 2.5 /CS clearly enhanced MMP-9 and MMP12 protein expression levels, while there was only a slight increase in the level of TGF-β1. Importantly, an evident enhancement in peri-bronchial fibronectin and collagen deposition were observed in PM 2.5 /CS. Moreover, PM 2.5 + CS led to an evident enhancement of these expressed protein levels.

Conclusions
In short, our data suggested that chronic prolonged PM 2.5 exposure induced impaired lung function, emphysematous lesions, airway inflammation and airway wall remolding. Most importantly, PM 2.5 exposure could also obviously aggravate CS-induced changes in COPD. This study offers a new insight into the effects induced by chronic PM 2.5 exposure and associated underlying mechanisms in COPD development and progression. However, it remains unclear whether these changes induced by PM 2.5 are its direct or indirect exposure effects. Further studies are required to extend these findings. The assessment of peribronchial collagen deposition in lung tissue induced by chronic exposure to PM 2.5 with or without CS. c Semiquantitative assessment of fibronectin protein expression using Image-Pro Plus. Results were expressed as means± SD (n = 6). *P < 0.05, ** P < 0.01, *** P < 0.001 vs control group; # P < 0.05, ## P < 0.01, ### P < 0.001 vs PM 2.5 + CSE-exposed group data acquisition in this study. JPZ, YX and XW contributed reagents, materials and analysis tools and provided help for analyzing the data. JW performed data analysis. JX conceived and designed the study, guided the methods of all the experiments, performed data analysis and revised the manuscript. All authors read and approved the final manuscript.