Environ. Sci. Technol. 2010, 44, 1424–1430
Cellular Responses after Exposure of Lung Cell Cultures to Secondary Organic Aerosol Particles A N N I N A G A S C H E N , †,# D O R I S L A N G , †,# M A R K U S K A L B E R E R , * ,§ M E L A N I E S A V I , † THOMAS GEISER,| AMIQ GAZDHAR,| CLAUS-MICHAEL LEHR,⊥ MICHAEL BUR,⊥ JOSEF DOMMEN,‡ URS BALTENSPERGER,‡ A N D M A R I A N N E G E I S E R * ,† Institute of Anatomy, University of Bern, 3012 Bern, Switzerland, Laboratory of Atmospheric Chemistry, Paul Scherrer Institut (PSI), 5232 Villigen, Switzerland, Centre for Atmospheric Science, Department of Chemistry, University of Cambridge, Cambridge CB2 1EW, U.K., Division of Pulmonary Medicine, University Hospital, 3010 Bern, Switzerland, and Department for Biopharmaceutics and Pharmaceutical Technology, University of Saarland, 66123 Saarbru ¨ cken, Germany
Received July 27, 2009. Revised manuscript received December 15, 2009. Accepted December 23, 2009.
The scope of this work was to examine in vitro responses of lung cells to secondary organic aerosol (SOA) particles, under realistic ambient air and physiological conditions occurring when particles are inhaled by mammals, using a novel particle deposition chamber. The cell cultures included cell types that are representative for the inner surface of airways and alveoli and are the target cells for inhaled particles. The results demonstrate that an exposure to SOA at ambient-air concentrations of about 104 particles/cm3 for 2 h leads to only moderate cellular responses. There is evidence for (i) cell type specific effects and for (ii) different effects of SOA originating from anthropogenic and biogenic precursors, i.e. 1,3,5trimethylbenzene (TMB) and R-pinene, respectively. There was no indication for cytotoxic effects but for subtle changes in cellular functions that are essential for lung homeostasis. Decreased phagocytic activity was found in human macrophages exposed to SOA from R-pinene. Alveolar epithelial wound repair was affected by TMB-SOA exposure, mainly because of altered cell spreading and migration at the edge of the wound. In addition, cellular responses were found to correlate with particle number concentration, as interleukin-8 production was increased in pig explants exposed to TMB-SOA with high particle numbers.
1. Introduction Ambient fine and ultrafine particles have, apart from impacts on atmospheric processes, a variety of adverse health effects from which respiratory diseases have attracted a lot of * Corresponding author
[email protected],
[email protected]. † University of Bern. ‡ Paul Scherrer Institut. § University of Cambridge. | University Hospital. ⊥ University of Saarland. # These authors contributed equally to this work. 1424
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 4, 2010
attention. A substantial fraction of inhaled particles is deposited in the lungs. Epidemiology has consistently shown correlations between morbidity and particles with diameter d e 2.5 µm (particulate matter, PM 2.5) (1). However, the interaction of particles with the inner lung surface, the main pathway of undesired particle uptake, is poorly understood (2). Organic compounds account often for 25-50% of the total particle mass. A part of the organic compounds found in particles is directly emitted into the atmosphere in particulate form whereas another fraction is formed in the atmosphere from gaseous precursors and is thus called secondary organic aerosol (SOA) particles, up to 70% of the nonrefractory submicrometer particle mass at various locations worldwide (3). SOA can have biogenic sources, i.e., emissions from trees and plants, or anthropogenic sources like traffic emissions or solvent evaporation. The chemical pathways involved in SOA formation include a large number of compounds. While 1,3,5-trimethylbenzene (TMB) is a typical precursor molecule representing SOA from anthropogenic sources, R-pinene is an important biogenic precursor (4). In most studies investigating particle-lung cell interactions in vitro, particle deposition differs greatly from the in vivo situation (5, 6), causing controversial results. Laboratory experiments investigating particle-lung interactions need to accurately reproduce the complex interaction between aerosol particles and the lung surface to identify the healthrelevant processes (7). We have previously presented a nanoparticle deposition chamber to expose lung cells to aerosol particles directly out of a conditioned air flow, therefore mimicking closely the particle deposition conditions in the lung (8). Experiments with inert polystyrene latex (PSL) particles showed that our newly designed aerosol deposition chamber is highly suitable for the exposure of lung cells to submicrometer particles. Particles are deposited evenly and efficiently at physiological conditions and a reliable quantification of the deposited particles is achieved. Control experiments illustrated that cells are not negatively affected by the handling procedure itself. All this is required when cell responses are to be quantified as a function of deposited nanoparticles. The inner lung surface, the primary target for inhaled particles, functions as a physical, biochemical, and immunological barrier to separate outside from inside. Its distinctive characteristics have to be included when studying mechanisms of particle-cell interactions in cell culture systems: (i) the epithelium consists of various highly differentiated and polarized cells with a low turnover; (ii) cells grow at the air-liquid interface and are thus in ultimate proximity to deposited particles; (iii) lung defense (particle clearance) is primarily provided by the mucociliary system and resident phagocytes, the macrophages. These factors were taken into account in the present project by working with different cell culture systems. This article describes biological results from the interdisciplinary POLYSOA project which includes the online application of SOA to representative lung cell culture systems for controlled studies of health effects by inhaled (nano)particles in vitro (9).
2. Methods 2.1. Experimental Layout. Cell cultures used in this study consisted of porcine and human lung epithelial cells (i.e., microdissected tracheal epithelia, primary cells, and cell lines) and of macrophages. Cells were cultured on microporous 10.1021/es902261m
2010 American Chemical Society
Published on Web 01/21/2010
TABLE 1. Overview of Experimental Conditionsa experiment
precursor type
precursor [ppb]
NO [ppb]
NO2 [ppb]
SO2 [ppb]
cell types
1 2 3 4 5 6 7 8 9
TMB TMB TMB TMB TMB TMB TMB R-pinene R-pinene
597 1144 1158 1225 1400 1360 1117 307 304
157 304 287 301 300b 300b 260 107 108
133 288 271 314 300b 300b 285 115 137
0.4 0.8 2 2 2 2 2 5.4 5
pig tracheal explants pig tracheal explants pig macrophages pig macrophages pig tracheal explants human and pig macrophages, pig tracheal epithelial cells A549 alveolar epithelial cells pig tracheal epithelial cells, pig macrophages human and pig macrophages
a Measured initial gas concentrations in the smog chamber and the cell types examined are listed. Note that different cell types (last column of table) were used during different experiments. The concentration ratios of NO2:NO:SOA precursor was about 1:1:4. b Values not measured due to device failure; target values are given instead.
filter inserts in multiwell plates. They were exposed to the aerosols at the air-liquid interface (ALI) in our newly developed particle deposition chamber (8). Cell cultures exposed to particle-filtered air from the smog chamber (“clean air control”) and untreated ones maintained in a regular tissue culture incubator (“incubator”) served as controls. The following biological end points were measured after aerosol exposure: • Cell and tissue integrity, assessed by transmission electron microscopy (TEM) at 24 h after aerosol exposure • Cytotoxic effects of aerosols, by measuring lactate dehydrogenase (LDH) activity released from damaged or lysed cells into the cell culture medium at 6 and 24 h after aerosol exposure • Clearance function of macrophages, by a 2 h postexposure treatment of macrophages with PSL particles and analysis of particle uptake using light microscopy (LM) and TEM • Production of inflammatory mediators, by measuring the release of tumor necrosis factor-R (TNF-R), interleukin-6 (IL-6), and IL-8 into the cell culture medium at 6 and 24 h after aerosol exposure • Alveolar epithelial repair function, by measuring the closure of mechanically wounded alveolar epithelial cell monolayers (A549) at 24 h after aerosol exposure, using a computerized imaging technique (10). 2.2. Generation of Aerosols. SOA were produced in the PSI indoor smog chamber as part of the POLYSOA project (see ref (9) for details). The smog chamber has been described in detail earlier (11). Briefly, it consists of a 27 m3 Teflon bag in a temperature-controlled housing. Aerosols were generated at 20 °C and 50% relative humidity. The chamber was first humidified before introducing NO, NO2, and the organic reactant (SOA precursor). TMB and R-pinene, model compounds for anthropogenic and biogenic precursors, respectively, were used to generate SOA. Four xenon arc lamps were used to simulate the solar light spectrum and to start photochemistry. Experiments with initial mixing ratios of about 600 or 1200 ppb TMB or 240 ppb R-pinene were performed. Organic precursor/NOx ratios of 2:1 were used for all experiments (Table 1). After 0.5-2 h aerosol particles formed in the smog chamber and rapidly increased in number to about 2 × 104 particles/cm3 in maximum and to 300 nm in size (Figures 1 and S1). Small amounts of SO2 were added to increase particle number concentration since SO2 is oxidized to H2SO4, which supports nucleation. Six hours after starting the experiment, about 0.02-0.1 µg/m3 H2SO4 particle mass is formed in the TMB experiments and 90% relative humidity, and 36 °C prior to contact with the cell cultures. Right before entering the deposition chamber, particles were charged with a bipolar Kr-85 charger, resulting in 0-4 net elemental charges per particle, which is similar to an atmospheric particle charge distribution. Particles were then deposited in an electrical field onto 6 or 12 individual cell cultures maintained at ALI at flow rates of 50 mL/min per cell culture. A deposition voltage of 4 kV/cm and alternating polarity (1 Hz) was applied, which resulted in a deposited particle fraction of about 20% of all particles as described earlier (8) in detail. In control experiments, cell cultures were exposed to clean air, i.e., smog chamber air with a Teflon particle filter inline to remove all SOA particles (labeled in figures as “clean air control”). This assures that experimental conditions are the same as for the particle-exposed cells and allows for determining the biological effects solely due to SOA particles. Cell cultures were exposed to SOA or to clean air for 2 h. 2.4. Cell Cultures. Cell cultures were prepared according to standard protocols; detailed information is provided in the Supporting Information. Cells were cultured at ALI except for the cell line and macrophages, which were submerged cell cultures. The basal medium of ALI cultures was changed daily and mucus was removed by washing the cell surfaces with phosphate-buffered saline (PBS, pH 7.4). For aerosol exposure at ALI, the apical medium of the submerged cell cultures was reduced to a minimum, i.e., to 0.2 mm in height. 2.5. Biological End Points. 2.5.1. Cell and Tissue Integrity. For ultratructural analysis, representative samples from SOAexposed and control cell cultures were fixed at 24 h after aerosol exposure in buffered 2.5% glutaraldehyde, 1% osmium tetroxide, and 0.5% uranyl actetate, dehydrated in ethanol, and embedded in Epon resin according to standard protocols in our laboratory (13, 14). Ultrathin sections of 60 nm thickness were cut and stained with lead citrate and uranyl acetate. Cells were examined according to predetermined criteria of assessment using a score sheet developed in our VOL. 44, NO. 4, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1425
FIGURE 1. Aerosol parameters and cell exposures. Time evolution of the ozone mixing ratio, the geometric mean diameter of the particles, and also particle number and particle mass concentrations during the experimental days 5 and 8. We assumed a density of 1 g/cm3 for the particles. The time spans of the three exposed cell series (particle-free air control [clean air control] and two subsequent experiments with exposures to SOA [aerosol A and aerosol B]) are indicated with shaded boxes. Figure S1 in the Supporting Information gives this data for all experiments. laboratory. Thereby, the integrity of cell contacts (junctional complexes), cell membranes, organelles, nuclei, and cytoplasm were evaluated. 2.5.2. Cytotoxicity. Cytotoxicity was assessed by assaying apical cell supernatants collected at 6 and 24 h postexposure for LDH activity released from the cytosol of damaged cells using the colorimetric Cytotoxicity Detection KitPLUS [LDH] (Roche Applied Science) according to the manufacturer’s protocol. Absorbance was measured in an enzyme-linked immunosorbent assay (ELISA) reader using the Magellan V6 Software (both TECAN Austria GmbH). Results were normalized to maximal releasable LDH by cell lysis. 2.5.3. Phagocytic Activity of Macrophages. To test the phagocytic activity of macrophages, 2 × 106/insert PSL particles (6 µm Fluoresbrite Polychromatic Red, Polysciences, Chemie Brunschwig AG, Basel, Switzerland) suspended in 1.5 mL of Dulbecco’s minimal essential medium were added to SOA-exposed and control macrophages immediately after the experiment for 2 h. Thereafter, the particle-containing medium was removed, cells were washed with PBS, and fresh medium was added. At 24 h after aerosol exposure, macrophages were fixed in buffered 2.5% glutaraldehyde, and the number of phagocytosed 6 µm PSL per macrophage was assessed by LM. On average, 102 (SEM 9) macrophages were counted on each sample. Particle internalization was confirmed by TEM. 2.5.4. Inflammatory Mediators. The release of the inflammatory mediators, IL-6, IL-8, and TNF-R, were assessed in apical cell supernatants collected at 6 and 24 h after aerosol exposure, using the Bio-Plex multiplex bead-based suspension array system and the appropriate detection kit (Bio-Rad Laboratories, California) according to the manufacturer’s protocol. As there were no ready-to-use beads commercially 1426
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 4, 2010
available for pig cytokines, pig antibodies were coupled to plain beads. Details are provided in the Supporting Information. 2.5.5. Alveolar Epithelial Wound Repair In Vitro. Alveolar epithelial repair activity was determined in an in vitro wound repair assay as previously described (15) using human A549 alveolar epithelial-like cells (American Type Culture Collection, Rockville, MD). Briefly, A549 cells were cultured to confluency on filter inserts in 6-well plates for 48 h and then mechanically wounded with a pipet tip. Wounded cell monolayers were exposed either to SOA or to clean air or were left untreated in the incubator. The area of the wound was measured over a time period of 24 h after aerosol exposure, and the rate of alveolar epithelial repair was expressed as percentage of reduction in wound area. In addition, to distinguish between the repair mechanisms, we assessed (i) cell spreading/migration by measuring the distances between cell nuclei at the wound edge and within the epithelial monolayer and (ii) cell proliferation by immunohistochemical detection of 5-bromo-2′-deoxyuridine (BrdU), incorporated in DNA during S (synthetic)-phase of the cell cycle, i.e., by determining the percentage of stained cell nuclei, as described previously (10). 2.5.6. Statistics. The standard error of the mean (SEM) was calculated by taking the ratio from the standard deviation (SD) and the square root of the sample number. The p-values were calculated using the standard tool from Microsoft Excel called “chi-test”. Values for p e 0.05 are assigned significant.
3. Results and Discussion 3.1. Aerosol. Aerosol data including time series of particle number, particle mass, particle diameter, and ozone concentrations for each experiment are shown in Figures 1 and S1. TMB was used as the precursor molecule in experiments
FIGURE 2. Phagocytic activity of macrophages. Phagocytic activity in pig (A, B) and human (C, D) macrophages exposed to SOA from TMB (A, C) and r-pinene (B, D). The number of phagocytosed PSL particles per macrophage (mph) was assessed by light microscopy; 102 (SEM 9) macrophages were evaluated per filter insert. The effect of SOA on particle uptake by pig macrophages was unclear because differences from controls were, though significant, either very small (A) or the differences applied to one of the controls only (B). Particle uptake in human macrophages was significantly impaired after exposure to particle-free air (clean air control, C, D) and to SOA from r-pinene (D) and “aged” TMB (C, aerosol-B). One bar represents the mean values and SD of three individual cell cultures. *, significant change compared to untreated cells, p e 0.05; #, significant change compared to clean air controls, p e 0.05. 1-7 and R-pinene in experiments 8 and 9. The concentration of TMB was about 1200 ppb except for experiment 1 with 600 ppb; that of R-pinene was about 300 ppb (Table 1). In general, particles formed more quickly when R-pinene was used as the precursor molecule, i.e., about 30 min after starting irradiation of the gas mixture in the smog chamber with light. Nucleation of TMB-SOA particles started 1.5-2 h after starting irradiation and reached the maximum particle number concentration about 1 h later for both precursor molecules. The maximum particle number ranged between 1 and 3 × 104/cm3. The maximum particle mass concentration was about 200 µg/m3 for TMB and about 300 µg/m3 for R-pinene. The maximum particle diameter was 400 nm for TMB and 300 nm for R-pinene. Ozone reached its maximum mixing ratio after 4 h (350 ppb, TMB) or 3 h (100 ppb, R-pinene), respectively. While the ozone mixing ratio remained stable for the R-pinene system, it started to decrease in TMB experiments. As mentioned in the Methods section, ozone was removed from the gas stream by activated charcoal denuders before exposure to the cells. 3.2. Cell and Tissue Integrity. Integrity of cells and tissues at the inner lung surface is crucial for organ function. Especially, damage of cell contacts, the cytoskeleton, and cellular membranes by inhaled particles have a high potential for acute adverse health effects. Overall, we found no evidence for ultrastructural alterations in cells and tissue exposed to SOA as compared to the controls. Occasional stress fibers and dilated endoplasmatic reticulum were registered, in aerosol exposed and control cultures. An example set of micrographs from microdissected tracheal explants is shown in Figure S2. 3.3. Cytotoxicity. The evaluation of LDH activity, released from the cytosol of damaged or lysed cells, is a precise and
commonly used test to determine the cytotoxic potential of compounds in environmental and medical research. In this study, we found no significant increase in LDH release in any of the SOA exposed cell cultures in comparison to the controls (data is not shown). Overall, LDH release in SOA exposed cells was
