A Novel Exposure System for the Efficient and Controlled Deposition

Jul 2, 2008 - Aerosol and Sensor Technology, University of Applied Sciences,. Northwestern Switzerland, Windisch, Switzerland. Received December 9 ...
0 downloads 0 Views 3MB Size
Environ. Sci. Technol. 2008, 42, 5667–5674

A Novel Exposure System for the Efficient and Controlled Deposition of Aerosol Particles onto Cell Cultures M E L A N I E S A V I , †,⊥ M A R K U S K A L B E R E R , * ,‡,⊥ D O R I S L A N G , † MANUEL RYSER,§ MARTIN FIERZ,| ˇ KA,§ ANNINA GASCHEN,‡ JAROSLAV RIC A N D M A R I A N N E G E I S E R * ,† Institute of Anatomy, University of Bern, Bern, Switzerland, Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich, Switzerland, Institute of Applied Physics, University of Bern, Bern, Switzerland, and Institute for Aerosol and Sensor Technology, University of Applied Sciences, Northwestern Switzerland, Windisch, Switzerland

Received December 9, 2007. Revised manuscript received March 20, 2008. Accepted May 7, 2008.

Epidemiologic studies have shown correlations between morbidity and particles e2.5 µm generated from pollution processesandmanufacturednanoparticles.Therebynanoparticles seem to play a specific role. The interaction of particles with the lung, the main pathway of undesired particle uptake, is poorly understood. In most studies investigating these interactions in vitro, particle deposition differs greatly from the in vivo situation, causing controversial results. We present a nanoparticle deposition chamber to expose lung cells mimicking closely the particle deposition conditions in the lung. In this new deposition chamber, particles are deposited very efficiently, reproducibly, and uniformly onto the cell culture, a key aspect if cell responses are quantified in respect to the deposited particle number. In situ analyses of the lung cells, e.g., the ciliary beat frequency, indicative of the defense capability of the cells, are complemented by off-line biochemical, physiological, and morphological cell analyses.

1. Introduction Adverse health effects associated with exposure to fine and ultrafine particles in ambient air are well documented by epidemiology (1–3) and supported by mechanistic animal studies (4–6). The particle properties responsible for these effects are only poorly understood, but particle size, surface area, and chemistry are likely important parameters (7–12). To elucidate the features of particle-cell interactions in cell culture systems representing the respiratory surface, it is essential to mimic the in vivo interactions of particles with cells as closely as possible. Because it is unknown which particle properties are causing negative biological effects, it is crucial that aerosol particles maintain their chemical and physical characteristics when they come into contact with * Author correspondence to either author. E-mail: kalberer@ org.chem.ethz.ch (M.K.); [email protected] (M.G.). † Institute of Anatomy, University of Bern. ‡ ETH Zurich. § Institute of Applied Physics, University of Bern. | University of Applied Sciences. ⊥ These authors contributed equally to this work 10.1021/es703075q CCC: $40.75

Published on Web 07/02/2008

 2008 American Chemical Society

the cells. However, commonly used model systems for in vitro studies deviate significantly from the in vivo situation such that responses are possibly not representative of those induced by the particles in vivo (13–16). Particle extracts or suspensions are often pipetted (14, 15, 17, 18), or sprayed (16) onto cell cultures. Additionally, with these exposure techniques cells are exposed to all the particles at once, resulting in very high particle deposition per time unit. Another approach is the CULTEX-System (19) where particles are deposited from a continuous aerosol flow by diffusional and gravitational deposition onto cells. However, these deposition processes are highly inefficient. Only 0.6% of suspended particles with 200 nm diameter are deposited in such a system (20). Because of this inefficient deposition only high-concentration particle sources can be investigated. Aerosols with lower particle concentrations e.g., ambient air, usually with 103-104 particles/cm3, either cannot be studied or necessitate excessively long exposure times. Consequently, ongoing cellular responses will overlap, complicating the results. Additionally, the maintenance of constant exposure conditions over a long time period is difficult and imposition of cellular stress is likely occurring. In most exposure systems used so far, the number and distribution of deposited particles per cell are insufficiently controlled. To evaluate the toxicity of particles, it is essential to quantify these parameters. Only in one recent study by Bitterle et al. (21) a stagnation point flow system was described, however, with a rather low particle deposition efficiency of only about 2%.

2. Experimental Section 2.1. Deposition Chamber. In the new deposition chamber described here, particles are deposited onto the cells directly out of a conditioned air-flow, simulating accurately the physiological conditions in the lung (Figure 1a). This assures that all cellular reactions are due to particle deposition and not due to other stress factors. One of the most important features of the chamber is the electrostatic particle deposition. Before entering the deposition chamber, particles pass a bipolar Kr-85 charger where they reach an equilibrium charge distribution. After the charger, 41-70% of all particles in the size range of 50-600 nm carry one to about five positive or negative net elemental charges (approximately equal positive and negative charges), whereas the rest is uncharged (22). Thus, the particles are not highly charged on their surface as would be the case for other charging processes, e.g., in corona chargers. Aerosol particles in the ambient atmosphere show a charge distribution, which is often very close to the charge equilibrium generated by the Kr-85 charger (23). After the Kr-85 charger, only copper tubing was used. After adjusting the gas composition to physiological conditions (i.e., 75% N2, 20% O2, 5% CO2) the aerosol is heated to 36 °C and humidified to 85-95% relative humidity. Humidity and temperature were measured continuously inside the aerosol particle deposition chamber during exposure. Particles enter the deposition chamber via a manifold (not shown in Figure 1a), which is a tube of 3.5 cm length and 2.7 cm diameter. At the bottom of the manifold six particle-delivery tubes are attached in a circular arrangement. Particles are directed individually to one of six cell cultures grown on filter inserts (24 mm diameter) in multiwell plates (Figure 1b). The delivery tubes have an inner diameter of 4 mm and increase gradually in diameter. For the last 25 mm, the delivery tubes have a cylindrical shape with an inner diameter of 18 mm. The exit tubes of the deposition chamber VOL. 42, NO. 15, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5667

FIGURE 1. Schematic of the experimental setup. (a) Overview of the experimental setup for online particle production, conditioning, and deposition onto cell cultures. Before entering the particle deposition chamber, the aerosol flow is conditioned to physiological requirements (i.e., humidification, 36 °C, 5% CO2) and particles are charged with a Kr-85 charger to enable the efficient electrostatic deposition. (b) Details of the particle deposition chamber showing the particle deposition on a cell culture filter insert inside the chamber. Six cell cultures are simultaneously and individually exposed to particles. Particles are deposited in an electrical field between the particle delivery tube and the electrode (20 mm diameter). A separate electrode is placed directly beneath each filter insert (distance from end of the delivery tube to electrode: 10 mm). The distance between the end of the delivery tube and the cell culture is 4 mm. have an inner diameter of 4 mm. The aerosol flow rate is 50 mL min-1 per filter insert, which is equal to 11 mL min-1 cm-2. The entire deposition chamber has a base area of 15 × 17 cm and a total height of 16 cm. Particles are deposited by an alternating, square-wave electrical field of 4 kV/cm applied between the end of each particle delivery tube and a circular electrode placed directly beneath each filter insert, assuring that the majority of the charged particles are deposited. The delivery tubes are equipped at the end with fine metal meshes (acting as grounded counter electrode) to ensure parallel electrical field lines between the tube end and the cell culture, and thus they allow for an even particle deposition on the cell cultures (see Figure 2b, red lines). To avoid accumulation of particles with one polarity on the cells, the polarity of the electrical field alternated at 1 Hz. Unipolar charging, e.g., with a corona charger, was not considered although it would be a more efficient charging process because bipolar charging and deposition mimics more closely the situation in the ambient atmosphere and the lung. Additionally, unipolar charged particles would require a constant electric field of one 5668

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 15, 2008

polarity applied to the electrodes, which could potentially induce undesired charge separation of ionic components in the culture medium below the lungs cells. The deposition of the charged particles results in a negligibly low electrical current on the cell cultures. At typical particle concentrations (104 particles/cm3) about 2000 particles/s (equal to about 3 × 106 particles/cm2 in 2 h) are deposited on an insert resulting in a current of about 0.3 fA. Cell damages as they are generated intentionally, e.g., in electroporation experiments, occur only at much higher currents of typically about 100 A (1017 times larger current). Therefore, the deposition of the charged particles used here is not expected to damage the cells. For online monitoring of the state of the cell culture, the chamber is equipped with a device detecting the mucociliary activity, whose most conspicuous signature is the ciliary beat frequency (CBF). A fiber optic dynamic light scattering probe (24) placed at each particle-delivery tube (Figure 1b) acts as an optical bistatic radar and detects the vertical movement of the mucus induced by the rhythmically beating cilia (generating metachronal waves) within the illuminated and

FIGURE 2. Distribution of deposited particles on a filter insert. (a) 200 nm PSP are evenly distributed along the filter insert radius over the entire filter except at the outer about 2 mm where significantly fewer particles were counted (average over 18 filter inserts and standard deviation). Two to three experiments were performed for each of the six inserts. The reason for the lower deposition at the filter edges is explained in the text. (b) Model calculation of the particle deposition on a filter insert. Particle trajectories are shown as white lines, and the color indicates the gas flow rate with higher flows in red and lower values in blue. Equipotential lines of the electric potential are drawn in red. Black lines indicate the solid surfaces of the aerosol delivery tube and filter insert. Most of the particles starting at the inner part of the tube are deposited, and only particles with trajectories at the edge of the delivery tube, where the flow rate is low, escape deposition. observed area with a high sensitivity (24). Two bare singlemode optical fibers are arranged approximately in specular reflection geometry. One of the fibers (Spectran Specialty Optics Co., SMC-A0780B-AO1VS, Avon, CT) delivers the illuminating beam from a diode-laser source (PL780T Laser Components, Pmax ) 25 mW, λ ) 785), and the other collects the scattered light within the numerical aperture of 0.1. The overlap of the illuminating beam and of the receiving beam defines a measuring area of about 20 mm2. The scattered signal is detected by a single-photon counter (SPCM-AQ, PerkinElmer), recorded with a multichannel scaler (FAST ComTec 7882 MCD-2, Oberhaching, Germany) and analyzed in terms of the power spectrum. The CBF is identified as the position of the fundamental harmonic of the power spectrum (see inset, Figure 4). 2.2. Test Particles. Test experiments to characterize the chamber were performed with monodisperse uncoated polystyrene particles in the size range of 50-600 nm (PSP, Duke Scientific Co., Palo Alto CA, 200 nm fluorescent PSP Fluoresbrite yellow/green from Polysciences Inc., Warrington

PA). PSP of 50, 100, 150, 200, 300, 400, 600 nm diameter were diluted with ultrapure water (18MΩ cm specific resistance) and nebulized with a home-built nebulizer operated with a critical orifice. The particle flow of 2 L per min passed through a silica gel diffusion dryer of 1 m length to ensure that water associated with the particles evaporated (relative humidity after the dryer: ca. 20%). The number concentration of PSP, measured continuously with an SMPS (DMA 3081, CPC 3022A, TSI, St. Paul, MN), was about 104 particles/cm3 for all particle sizes investigated. These particle concentrations are comparable to ambient-air concentrations (25). 2.3. Cell Cultures. We tested the new deposition chamber, except for ciliary beat frequency measurements, on cell cultures of a human bronchial-epithelial cell line (BEAS-2B) and on cultures of porcine lung macrophages. Macrophages were obtained by bronchoalveolar lavage (BAL) of porcine lungs. One lung lobe was flushed 10 times with 50 mL each of phosphate-buffered saline (PBS, pH 7.4), the recovered fluid centrifuged, and the cell pellet resuspended in 40 mL of PBS. The obtained 5 to 8 × 107 macrophages were cultured in petri dishes in minimal essential medium (D-MEM high glucose, Gibco, Lubioscience, Lucerne, Switzerland) for 1 h and then frozen in liquid nitrogen if not used immediately. The BEAS-2B cells were cultured in D-MEM low glucose (Gibco). Both culture media were supplemented with 10% fetal calf serum (FCS, Amimed, Allschwil, Switzerland) and 1% penicillin-streptomycin (Gibco). Cells were cultured at 37 °C, 90% relative humidity, and 5% CO2. Two days prior to aerosol exposure, cells were seeded at a density of 3 × 106 (macrophages) and 2 × 105 (BEAS-2B) cells per filter insert (FALCON, 24 mm diameter, 0.4 µm pore size, PET track-etched membrane, Milian, Geneva, Switzerland) in six-well plates (FALCON, Milian). BEAS-2B cells reached a confluency of >80% on the experimental day. For particle exposure, the cultures of BEAS-2B cells as well as those of macrophages were covered with a minimal amount of cell culture medium (