Subjects and bronchoalveolar lavage (BAL) collection
The study was approved by the Taipei Medical University-Joint Institutional Review Board (TMU-JIRB no. 201310027), and informed consent was obtained from all human subjects. In total, 47 healthy subjects (female: male of 24:23) with an average age of 53 years were recruited for the study. All subjects underwent a physical examination, including a pulmonary function test and blood collection. None of the subjects had a history of allergies or had suffered an airway infection within 3 months before recruitment into the study. The pulmonary function test was conducted immediately before bronchoscopy and included parameters of the forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1), taken with a spirometer. Bronchoscopy was performed with a flexible fiberoptic bronchoscope under local anesthesia. A bronchial wash was performed by instilling 100 ml of normal saline at 37 °C into the proper lung segment, and then the fluid was gently aspirated. Bronchoalveolar lavage fluid (BAL) was thus collected. The BAL was filtered through a single layer of loose sterile surgical gauze and collected followed by centrifugation at 400 ×g for 10 min at 4 °C, and the supernatant was collected for biochemical analysis.
Air pollution exposure
Exposure to PM2.5, NO2, and O3 was estimated using a hybrid kriging-land-use regression (LUR) approach as described in previous reports [18-20]. Briefly, in situ observations of PM2.5, NO2, and O3 were collected from Taiwan EPA air quality monitoring stations (https://airtw.epa.gov.tw/ENG/default.aspx). Predictor variables with an absolute value of a Spearman correlation coefficient of >0.4 with a prior direction of effect on the air pollution concentration were maintained and entered into stepwise linear regression procedures. Next, a set of interpolated pollutant levels were generated through a leave-one-out ordinary kriging function and added into the LUR model as a variable to improve the model performance. We obtained adjusted model R2 values of 0.89 for PM2.5, 0.87 for NO2, and 0.74 for O3 using the resultant hybrid kriging-LUR models. The models provided reliable air pollution exposure assessments in all subjects.
Chronic exposure to air pollution in vivo
The animal study was approved by the Animal and Ethics Review Committee of the Laboratory Animal Center at Taipei Medical University (no. LAC-2015-0290). Six-week-old male Sprague-Dawley (SD) rats obtained from the National Laboratory Animal Center (Taipei, Taiwan) were exposed to traffic-dominated urban air pollution. Rats were whole-body exposed to (1) clean air in the Laboratory Animal Center of Taipei Medical University (as the control group), (2) high-efficiency particulate air (HEPA)-filtered traffic-related urban air (as the HEPA group), and (3) traffic-dominated urban PM2.5 (as the PM2.5 group) for 3 and 6 months. The whole-body exposure system to unconcentrated particulate pollution with a 2.5-μm particle size classifier for rodents was previously described [21, 22]. Gaseous pollutants used for rat exposure were referenced from the nearest Yonghe EPA air quality monitoring station. After 3 and 6 months of exposure, rats were euthanized followed by collection of BAL, lung tissues, and serum (n=8 in each group).
In vitro experiment
Human lung alveolar epithelial A549 cells obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) were cultured in RPMI medium containing 10% fetal bovine serum, penicillin, and streptomycin. Cells were incubated at 37 C with 95% humidity and 5% CO2. Standard Reference Material (SRM) diesel exhaust particles (DEPs) obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA) were used to expose cells at 0 (control), 50, and 100 μg/mL at 37 C for 24 h in a humidified atmosphere with 5% CO2 (n=6 per group). Supernatants and cell lysates were collected for biochemical analyses.
Enzyme-linked immunosorbent assay (ELISA)
An ELISA was used to determine levels of 8-isoprostane (Cayman, Ann Arbor, MI, USA) and IL-6 (R&D Systems, Minneapolis, MN, USA) in BAL of humans and rats and in cell supernatants. ITIH4 in human BAL, rat BAL, lung lysates, and serum was examined using an ELISA (MyBioSource, San Diego, CA, USA). Details of the analytical procedures followed the manufacturer’s instructions. Results obtained from lung lysates are presented after adjusting for the total protein.
Immunohistochemistry (IHC)
Lung tissues of rats after 6 months of exposure were fixed in 2% paraformaldehyde followed by permeabilization with 0.1% Triton X-100 in 0.01 M phosphate-buffered saline (PBS; pH 7.4; containing 0.2% bovine serum albumin). Lung sections were then incubated with a polyclonal antibody against ITIH4 (1:1000) in PBS containing 3% normal goat serum, whereas incubation with PBS served as a negative control. An anti-rabbit immunoglobulin G (IgG) FITC-conjugated secondary antibody (Jackson ImmunoResearch, PA, USA; 1:500 dilution) was incubated with lung sections, followed by staining with 4’,6-diamidino-2-phenylindole (DAPI). Microphotographs were acquired using the Motic Easyscan Pro and the Motic DSAssistant software (Motic, Hong Kong, China).
Western blotting
Expressions of ITIH4, sirtuin 1 (Sirt1), phosphorylated extracellular signal-regulated kinase (p-ERK), caspase-3, matrix metallopeptidase 9 (MMP9), and MMP12 were examined using Western blotting as described in our previous report [23]. Briefly, lysate samples were electrotransferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Darmstadt, Germany). The primary antibodies for ITIH4 (1:1000), Sirt1 (1:1000), p-ERK (1:1000), caspase-3 (1:1000), MMP9 (1:1000), MMP12 (1:1000), and α-tubulin (1:1000) were obtained from Cell Signaling (Danvers, MA, US). Anti-rabbit (1:2000) horseradish peroxidase (HRP)-conjugated secondary antibodies were obtained from Chemicon International (MA, USA) and Merck Millipore (MA, USA). An HRP-labeled secondary antibody was incubated and washed with TBST after blocking. Enhanced chemiluminescence Western blotting reagents were used. Images were taken using the ChemiDoc MP imager (Bio-Rad, Hercules, CA, USA). Quantitative data were obtained using Image-Pro vers. 4 (Media Cybernetics, MD, USA) for Windows. All data were adjusted to the control (multiples of change of the control) as previously reported [24].
Statistical analysis
We applied a linear regression model to examine associations between air pollutants and BAL biomarkers in healthy subjects. Exposure variables were 1-year average PM2.5, NO2, and O3, and dependent variables were ITIH4, 8-isoprostane, and IL-6. Each regression model included age, sex, and current smoker (yes or no). The effects of air pollutants on BAL biomarkers are expressed as unit changes (β) multiplied by the interquartile range (IQR). A nonparametric Kruskal-Wallis test with Dunn's post-hoc test was used to compare multiple variables. Pearson's correlation was used to determine correlations of lung ITIH4 with BAL ITIH4 and serum ITIH4. All of the statistical analyses in this study were conducted with SPSS 15.0 software (SPSS, New York, NY, USA). The level of significance for all of the statistical analyses was set to p<0.05.