- Email: [email protected]
In vitro study of the pulmonary translocation of nanoparticles
Toxicology Letters 160 (2006) 218–226
In vitro study of the pulmonary translocation of nanoparticles A preliminary study J. Geys a , L. Coenegrachts a , J. Vercammen b , Y. Engelborghs b , A. Nemmar a , B. Nemery a , P.H.M. Hoet a,∗ a
Laboratory of Pneumology, Unit of Lung Toxicology, K.U. Leuven, Herestraat 49, Leuven 3000, Belgium b Laboratory of Biomolecular Dynamics, K.U. Leuven, Celestijnenlaan 200D, Leuven 3001, Belgium Received 20 May 2005; received in revised form 13 July 2005; accepted 13 July 2005 Available online 30 August 2005
Abstract Recent studies indicate that inhaled ultrafine particles can pass into the circulation. To study this translocation in an in vitro model three types of pulmonary epithelial cells were examined. The integrity of the cell monolayer was verified by measuring the transepithelial electrical resistance (TEER) and passage of sodium fluorescein. TEER was too low in A549 cells. In these preliminary experiments, TEER values of 1007 ± 300 and 348 ± 62 cm2 were reached for the Calu-3 cell line, using permeable membranes of 0.4 and 3 m pore size, respectively. Growing primary rat type II pneumocytes on 0.4 m pores, a TEER value of 241 ± 90 cm2 was reached on day 5; on 3 m pores, no acceptable high TEER value was obtained. Translocation studies were done using 46 nm fluorescent polystyrene particles. When incubating polystyrene particles on membranes without a cellular monolayer, significant translocation was only observed using 3 m pores: 67.5% and 52.7% for carboxyl- and amine-modified particles, respectively. Only the Calu-3 cell line was used in an initial experiment to investigate the translocation: on 0.4 m pores no translocation was observed, on 3 m pores ∼6% translocation was observed both for carboxyl- and amine-modified particles. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Rat type II pneumocytes; Calu-3 cell line; In vitro; TEER; Nanoparticles
1. Introduction Because of their physico-chemical properties, nanostructured materials (<0.1 m) can significantly differ from the corresponding coarse bulk material. The dosimetric aspects of the deposition and disposition of particles are different for inhaled ultrafine particles compared with larger particles (Oberd¨orster, 2001). Nanomaterials potentially have important industrial advantages. Concerns have been expressed that these properties also
∗
Corresponding author. E-mail address: [email protected] (P.H.M. Hoet).
might have adverse health and environmental impacts (Donaldson et al., 2004; Colvin, 2003; Hoet et al., 2004). The effect of nanostructured materials on biological systems has been studied mainly in view of the health risks posed by inhaling particles originating from air pollution: PM10 (particulate matter <10 m aerodynamic diameter), PM2.5 and PM0.1 (ultrafine particles ‘UFP’, diameter <0.1 m). Recent studies indicate that ultrafine particles can pass into the circulation (Nemmar et al., 2001). Although still somewhat controversial, it is accepted that some types of ultrafine particulates, such as elemental carbon, can be translocated to the blood circulation with subsequent uptake in the liver by 1 day after inhalation
0378-4274/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2005.07.005
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
exposure in rats (Oberd¨orster et al., 2002). Nemmar et al. (2002a, 2004) also showed translocation of ultrafine 99m Tc-carbon particles to the blood in healthy human volunteers. Recently, Oberd¨orster et al. (2004) demonstrated translocation of ultrafine particles to the brain via the olfactory nerve. The physico-chemical properties of the particulate surface play an important role in the biological effects in the lung and systemic circulation (Hamoir et al., 2003). Unmodified and negatively charged UFPs had no effect on thrombus formation whereas positively charged UFPs enhanced thrombus formation when administrated intravenously (Nemmar et al., 2002b) or instilled intratracheally (Nemmar et al., 2003). The Royal Society and Royal Academy of Engineering (RS/RAE 2004) recommend that industry should assess the risk of releasing products and materials containing nanoparticles or nanotubes throughout the lifecycle of the product and make this information available to the relevant regulatory authorities. The potential hazard of ultrafine particles and fibres is not predictable ‘a priori’ by their bulk physicochemical properties. Because in vivo experiments are expensive, slow and ethically questionable there is a strong demand for low-cost high-throughput in vitro assays without reducing the efficiency and reliability of the risk assessment (Luther et al., 2004). An in vitro cell culture model of the respiratory epithelium that closely mimics the human respiratory epithelium would be a valuable tool to screen ultrafine particles (or nanoparticles) for their capability to translocate through the pulmonary epithelium and to study this phenomenon in detail. Several cell lines are currently used as in vitro models of respiratory epithelium, e.g. A549 and Calu-3. The main purpose of this article is: (a) to assess the suitability of an in vitro model for studying the translocation of ultrafine particles across the respiratory epithelium and (b) to present preliminary data on the translocation of ultrafine model particles with different surface chemistry.
219
polystyrene carboxyl-modified fluorescent labelled particles (L5030, L3030) and sodium fluorescein were purchased from Sigma NV/SA (Bornem, Belgium). Deoxyribonucleic (DNase) I was obtained from Roche Molecular Biochemicals (Mannheim, Germany); Vitrogen collagen (3.0 mg/ml) from Nutacon (Leimuiden, The Netherlands) and Nembutal® (pentobarbital 60 mg/ml) from CEVA Sant´e Animal (Brussels, Belgium). Culture flasks were purchased from Iwaki Glass, International Medical (Brussels, Belgium); nylon filters (40 m) from Falcon, Becton Dickinson (Leuven, Belgium). Transwell Polycarbonate Membrane® and Transwell-Clear Polyester Membrane® inserts were obtained from Costar Europe (The Netherlands). The components of phosphate buffered saline (PBS− ) are 130 mM NaCl; 5.2 mM KCl; 10 mM glucose; 10.6 mM Hepes and 2.6 mM Na2 HPO4 ; pH 7.4; PBS+ contains in addition: 1.9 mM CaCl2 and 1.29 mM MgSO4 ; “non-fluorescent buffer” is composed of 118 mM NaCl; 4.75 mM KCl; 2.53 mM CaCl2 ·2H2 O; 2.44 mM MgSO4 ; 1.19 mM KH2 PO4 ; 25 mM NaHCO3 (Hamoir et al., 2003). 2.2. Cell lines
2. Materials and methods
The A549 (human alveolar epithelial cell line) and Calu-3 (human bronchial epithelial cell line) cell lines were obtained from ATCC and maintained as a monolayer culture in plastic tissue culture flasks. For the A549 cell line a Waymouth medium with 10% fetal calf serum, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM lglutamine and 1.25 g/ml fungizone was used. The Calu-3 cell line was maintained in MEM with 10% fetal calf serum, 1% non-essential amino acids, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM l-glutamine and 1.25 g/ml fungizone. The cells were incubated at 37 ◦ C, in a 100% humidified atmosphere containing 5% CO2 . When confluent, the cell lines were released enzymatically (0.1% trypsin–EDTA in PBS− ) and transferred to different culture plates. The medium was renewed every 2–3 days.
2.1. Materials
2.3. Isolation of primary type II pneumocytes
Waymouth medium, minimal essential medium (MEM), fetal calf serum, fungizone, l-glutamine (200 mM), penicillin–streptomycin (10,000 U/ml and 10,000 g/ml), non-essential amino acids and trypsin–EDTA were obtained from Invitrogen (Merelbeke, Belgium). Trypsin type I (T-8003), bovine serum albumin (BSA), Percoll, polystyrene amine-modified fluorescent labelled particles (L0780, L9529, L0155),
The method of Hoet et al. (1995) was used to isolate primary rat type II pneumocytes. Briefly, 150–200 g, male Wistar rats were anaesthetized with an intraperitoneal injection of Nembutal® (pentobarbital 1.5 ml/kg). The rats were exsanguinated and then tracheostomized. The lungs were perfused with 0.9% NaCl via the pulmonary artery. Finally, the heart was removed and the lungs together with the trachea were cut off. The lungs
220
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
were lavaged five times with 5–6 ml of 0.9% NaCl and then once with trypsin (250 mg in 100 ml PBS+ , pH 7.4). Then the lungs were filled every 5 min with trypsin and incubated at 37 ◦ C for 30 min. The trachea and bronchi were removed and the lungs were cut for 5 min until fragments smaller than 1 mm3 were obtained. Fetal calf serum (5 ml) and DNase I (1 mg) were added and the suspension was shaked for 5 min. After filtering through gauze and a 40 m nylon filter, the cell suspension was centrifuged for 30 min at 250 × g on a discontinuous Percoll gradient (ρ = 1.089 and 1.040 g/ml). The band on the interface was collected and pelleted by centrifugation for 10 min at 250 × g. The pellet of cells was resuspended in Waymouth medium with 10% fetal calf serum, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM l-glutamine and 1.25 g/ml fungizone and two times incubated for 40 min (37 ◦ C, 5% CO2 ) on a 60 mm petri dish. After centrifuging the cells, they were resuspended in the medium and counted using a B¨urker chamber. 2.4. Cell culture The cells were seeded onto Transwell Clear Polyester inserts of 0.33 cm2 or 1 cm2 , with pores of 0.4 m or 3 m diameter. Some experiments were done with Transwell Polycarbonate Membrane. The inserts were used uncoated, coated with mono-fibrillar collagen (Vitrogen100 diluted five times in PBS+ dried overnight) or coated with collagen gel (Vitrogen-100, 10× PBS− , pH 7.4, 60 min gelation at 37 ◦ C). Different seeding densities were used: 0.075, 0.15, 0.30, 0.45, 0.60, 0.75 and 0.90 × 106 cells/cm2 for the A549 cell line; 0.30, 0.60 and 1.20 × 106 cells/cm2 for the Calu-3 cell line and 0.30, 0.60 and 0.90 × 106 cells/cm2 for the primary type II cells. The cell monolayer was microscopically examined. 2.5. Measuring the integrity of the monolayer 2.5.1. Measurement of transepithelial electrical resistance Transepithelial electrical resistance was measured, after refreshing the culture medium, using MillicellERS (Millipore, Bedford, MA). Each sample was measured twice and the mean value was calculated. The mean resistance of a cell-free Transwell insert was subtracted from the resistance measured across each cell layer to yield the TEER value of the cell monolayer. The TEER of the Calu-3 cell line was measured daily from day 3 after seeding the cell culture until day 7; the type II pneumocytes were checked on days 3 and 5.
2.5.2. Sodium fluorescein transport Sodium fluorescein (1 mg/ml in non-fluorescent buffer) was added in the apical compartment, nonfluorescent buffer in the basolateral compartment. After placing the cultures for 60 min in the CO2 incubator (5% CO2 , 37 ◦ C) samples were taken from the basolateral compartment and the amount of sodium fluorescein was measured by spectrophotometry at 486 nm. The concentration in the samples was calculated using a standard curve with known concentrations of sodium fluorescein (10, 5, 2.5, 1.25, 0.625 and 0.3125 g/ml). The sodium fluorescein leakage was measured on the last day of the culture. A monolayer was considered to be a “tight monolayer” when the amount of sodium fluorescein in the basolateral compartment was less than 1% of the initial amount in the apical compartment. 2.6. Translocation studies In the basolateral compartment, the medium was removed and the compartment was filled with nonfluorescent buffer with 1% bovine serum albumin. The apical medium was replaced by medium without serum, containing 1% BSA and particles (25 g/ml). Two kinds of fluorophore-labelled polystyrene particles were used: amine-modified and carboxyl-modified polystyrene beads with a diameter of 0.046 m. Prior to use, the particle stock solution was sonicated for 15 min in a sonication-bath (Branson 1200, Branson Ultrasonics Corporation) and vortexed to reduce nonspecific aggregation. Translocation studies with Calu3 cells (0.60 × 106 cells/cm2 ) were performed on day 3 (0.4 m pores) and day 5 (3 m pores). Translocation studies with the primary type II pneumocytes (0.60 × 106 cells/cm2 ) were done on 0.4 m pores, on day 5. After 14–16 h of incubation, the medium in the apical and basolateral compartment was collected and the concentration of particles was measured. The concentration of the particles in the apical and basolateral compartment was assessed by fluorescence correlation spectroscopy (FCS) using a Confocor I (Zeiss Gmbh, Jena, Germany) confocal microscope. The 488 nm line of the Argon ion laser (power of 10 mW) was used for excitation and the emission of the fluoro-labelled particles was detected with an Avalanche Photodiode. The pinhole diameter was 60 m and the instrument was calibrated with Rhodamine 6G. The fluorescent molecules were measured in an open confocal volume with a typical size of 1.4 fl. Measurements were performed 10 times during 60 s (sample rate = 5 ms). To analyse the fluorescence fluctuations the method of Van Craenenbroeck et al. (2001) was used. This method dis-
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
221
Fig. 1. Transepithelial electrical resistance of Calu-3 cell monolayers. Calu-3 cells are seeded on Transwell Clear inserts with (A) 0.4 m pore size or (B) 3 m pore size; 0.3 × 106 cells/cm2 () or 0.6 × 106 cells/cm2 (). Data represent the mean TEER values ± S.D.; n = 4–3; three to two experiments (0.3 × 106 cells/cm2 ); n = 10; four experiments (0.6 × 106 cells/cm2 ). Student t-test between different seeding densities at the same day after seeding. ** p < 0.01 and * p < 0.05.
criminates the fluorescence intensities originating from free fluorophores or from bright spikes, using the statistical software S-Plus (Mathsoft Inc.). The intensities of the peaks of fluorescence were counted and the mean and standard deviation of the 10 measurements was derived. The concentration of the particles was calculated using a standard curve (5, 1, 0.2, 0.04 g/ml; four measurements each). 2.7. Statistical analyses Results are expressed as mean ± S.D. The data were analysed with paired or unpaired t-test and one-way ANOVA followed by Tukey’s multiple-comparison test (GraphPad Prism Package, GraphPad Software, San Diego, CA). p < 0.05 are considered significant. 3. Results 3.1. In vitro model A549 cells were seeded at different seeding densities on membranes with pore diameters of 0.4 m (0.075, 0.15, 0.30, 0.45, 0.60, 0.75 and 0.90 × 106 cells/cm2 ) and 3 m (0.15 and 0.45 × 106 cells/cm2 ), with or without collagen mono-fibrillar coating (n = 2 in two experiments). In all conditions, the TEER values did not exceed 25 cm2 and the permeability of sodium fluorescein was the same as that of the cell-free control inserts (data not shown). Seeding Calu-3 cells on 0.4 m pore size Transwell Clear inserts, mean TEER values of 404 ± 83 and 1007 ± 300 cm2 were reached on the third day for seeding densities of 0.30 × 106 and 0.60 × 106 cells/cm2 , respectively (Fig. 1A). After day 3, the TEER value declined. Calu-3 cells seeded on 3 m pore size membranes reached the highest
TEER value on day 5 (0.60 × 106 cells/cm2 ) or day 7 (0.30 × 106 cells/cm2 ) (Fig. 1B). The yield of primary type II pneumocytes was 4.3 ± 2.7 × 106 cells/rat. The viability, assessed with trypan blue exclusion, was more than 98%, and the purity, before plating just after isolation, was more than 85%. The cells were seeded on day 0 and the TEER was measured from day 3 on. The measured TEER values with different experimental conditions are presented in Table 1. Cells seeded on the 0.4 m pores formed a monolayer with the highest TEER value on day 5 after seeding. On the membrane with 3 m pores, primary rat type II pneumocytes did not develop a “tight monolayer”. When using mono-fibrillar collagen coating, the TEER values were very low. Cells seeded on collagen gel did not form a monolayer, the TEER values were as low as the control inserts without a cell layer (data not shown). Some experiments were also done using Transwell Polycarbonate Membrane® inserts. The TEER values in these experiments were lower than the TEER values in experiments using Transwell Clear Polyester Membrane® inserts (data not shown). 3.2. Integrity of the monolayer 3.2.1. Sodium fluorescein leakage The permeability of the monolayers was checked using sodium fluorescein. Plotting the TEER value and the permeability of sodium fluorescein against each other on a logarithmic scale, a linear correlation was found, both for the type II pneumocytes and the Calu3 cell line. This correlation is shown in Fig. 2 for the type II pneumocytes. According to Agu et al. (2001), we considered a monolayer with a leakage of less than 1% sodium fluorescein as a “tight monolayer”. Using the correlation between sodium fluorescein leakage and TEER a minimum TEER value for a “tight
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
222 Table 1 TEER values of type II pneumocytes
Day 3
Day 5
69 ± 23 cm2 (n = 11/5) isolations 24 ± 15 cm2 (n = 10/5)
224 ± 51 cm2 (n = 11/5) 89 ± 49 cm2 (n = 10/5)
0.4 m, 0.60 × 106 cells/cm2 No coating Coatinga
78 ± 46 cm2 (n = 25/11) 40 ± 21 cm2 (n = 12/11)
241 ± 90 cm2 (n = 25/11) 168 ± 56 cm2 (n = 12/11)
3 m, 0.60 × 106 cells/cm2 No coating Coatinga
7 ± 7 cm2 (n = 8/4) 10 ± 7 cm2 (n = 8/4)
31 ± 13 cm2 (n = 8/4) 25 ± 10 cm2 (n = 8/4)
3 m, 0.90 × 106 cells/cm2 No coating Coatinga
8 ± 3 cm2 (n = 10/4) 13 ± 4 cm2 (n = 8/4)
24 ± 9 cm2 (n = 10/4) 30 ± 6 cm2 (n = 8/4)
0.30 × 106
0.4 m, No coating Coatinga
cells/cm2
n = #/# isolations. a Mono-fibrillar collagen coating.
Fig. 2. Relation permeability sodium fluorescein and TEER value. Type II pneumocytes are seeded on 0.4 m pore size Transwell Clear inserts without mono-fibrillar collagen coating; n = 37. (A) Relation between permeability sodium fluorescein and TEER values. (B) Logarithmic transformation of both parameters and linear regression (r2 = 0.89; p < 0.0001). The line points to a leakage of 1% sodium fluorescein, with which the corresponding TEER value can be calculated.
monolayer” was calculated. The minimal TEER values for Calu-3 cells seeded on 0.4 and 3 m pores were 575 and 420 cm2 , respectively. The minimal TEER value of primary type II pneumocytes seeded on 0.4 m pores without mono-fibrillar collagen coating was 251 cm2 and with mono-fibrillar collagen coating 304 cm2 . 3.3. Translocation studies 3.3.1. Influence of changing the cell culture medium on TEER Changing the culture medium of the primary type II cells to non-fluorescent buffer had no effect on the TEER value (n = 5, see Fig. 3C). Incubating the culture with carboxyl-modified or amine-modified particles lowered
the TEER value of primary type II cells, but not significantly (n = 5, see Fig. 3C). After 14–16 h of incubation of the Calu-3 monolayer with particles, the TEER values were significantly higher under all conditions compared to before incubation (n = 6, see Fig. 3A and B). 3.3.2. Translocation through inserts without cells The distribution of the particles, after 14–16 h incubation, between the apical and the basolateral compartment of an insert cell culture system was first examined without the presence of a cell monolayer. The results are shown in Fig. 4. The translocation of carboxyl-modified particles was 13.5% through 0.4 m pores (n = 7) and 67.5% through 3 m pores (n = 3). The translocation of amine-modified particles was 4.2% through 0.4 m pores (n = 7) and 52.7% through 3 m pores (n = 3).
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
223
Fig. 3. Influence of changing the cell culture medium on TEER. Calu-3 cells are seeded on a (A) 0.4 m pores (n = 6) and (B) 3 m pores (n = 6); primary type II pneumocytes are seeded on 0.4 m pores (C) (n = 5). Cell monolayers are incubated with 25 g/ml carboxyl-modified particles () or amine-modified particles (), or non-fluorescent buffer (䊉). Data represent mean ± S.D. *** p < 0.001, ** p < 0.01 and * p < 0.05.
3.3.3. Translocation through a cell monolayer Because the TEER values of the A549 culture were too low, no translocation studies were done with this cell line. The translocation studies were performed using the Calu-3 cell line (0.4 and 3 m pore size) and primary type II cells (0.4 m pore size). Incubating Calu-3 cell monolayers with particles (25 g/ml) on the 0.4 m pore size membrane, we could not detect particles in the basolateral compartment. Fig. 5 shows the individual data of six different cultures of Calu-3 cell monolayers growing on 3 m pore size membranes incubated overnight with 46 nm particles. A limited fraction of the particles, both carboxyl- or amine-modified particles (25 g/ml), incubated at the apical side of the cultures translocated to the basolateral compartment. We could not detect
any translocation in 2 (of 5) and 3 (of 6) experiments with carboxyl- or amine-modified particles, respectively. After incubation, 1.05 (±0.29) g particles and 1.16 (±0.19) g particles were in the apical compartment of the carboxyl-modified particles and amine-modified particles, respectively. In the basolateral compartment 0.06 (±0.05) g particles were found for both particle types. This represents, on average, a translocation of ∼6% of the particles from the apical to the basolateral compartment. After incubating the primary type II pneumocytes monolayer with particles on the 0.4 m pore size membrane, we could not detect any translocation to the basolateral compartment (data not shown). Because of too low TEER values no translocation studies were done with 3 m pore size.
Fig. 4. Translocation of particles through inserts without a cell monolayer. Carboxyl-modified or amine-modified particles were introduced at the apical compartment on inserts without a cell monolayer. The concentration of particles in both compartments after 14–16 h incubation is measured. (A) carboxyl-modified particles; 0.4 m pore size, n = 7; (B) carboxyl-modified particles; 3 m, n = 3; (C) amine-modified particles; 0.4 m, n = 7; (D) amine-modified particles, 3 m, n = 3. Data represent mean ± S.D.
224
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
Fig. 5. Translocation of particles through a Calu-3 monolayer. Calu-3 cells are seeded on a 3 m pore membrane and incubated with 25 g/ml carboxyl-modified particles (A) or amine-modified particles (B). Per monolayer culture three values are given: the first two (connected with full line) present the TEER value before and after the incubation with particles. These two values have to be read from the left y-axis. The third value (most right), presents the total (g) mass of particles translocated into the basolateral compartment (n = 6).
4. Discussion Recent studies indicate that inhaled ultrafine particles can pass into the circulation (Nemmar et al., 2001, 2002a; Oberd¨orster et al., 2002). Moreover, the physicochemical properties of the particulate surface plays an important role in the biological effects in the lung and systemic circulation (Hamoir et al., 2003). At this moment the mechanisms and kinetics of particle translocation are not yet known. Therefore, the aim of this study was to build an in vitro model, including the setup of a suitable cell culture system, to study this translocation phenomenon. This can be a tool in screening the toxicity and bioavailability of different materials before marketing. Moreover, such a model can also be a valuable tool for pharmacological studies. 4.1. Integrity of cell monolayer In developing an in vitro model it is important that it reflects the in vivo situation as closely as possible. At this moment such a system has not yet been described. We evaluated three types of cell monolayers to mimic the in vivo respiratory epithelial barrier: the commercially available and widely used A549 and Calu-3 cell lines, as well as primary rat type II pneumocytes, which are more cumbersome to obtain. In general, a monolayer is said to be a “tight monolayer” if the leakage of sodium fluorescein is less than 1% (Agu et al., 2001). However, the disadvantage of sodium fluorescein is the inability to use the culture afterwards. We noted a good correlation between the TEER value and the leakage of sodium fluorescein. Thanks to this correlation it is possible to monitor the tightness of the cell layer, using TEER, before (and after) use in an experiment.
In our experiments, we were unable to obtain high TEER values with the A549 cell line. This is consistent with the reported inability of the A549 cell line to form tight junctions (Winton et al., 1998). Also Foster et al. (1998, 2001) reported low TEER values using the A549 cell line. This indicates that the A549 cell line is not suitable for studying translocation. The present study confirms previous reports indicating that the Calu-3 cell line is capable of forming tight monolayers in culture, thus generating high TEER values. Working with the Calu-3 cell line, TEER values of 1007 ± 300 and 348 ± 62 cm2 were obtained using permeable membranes with 0.4 and 3 m pore size, respectively. A large pore size reduces significantly the capacity of the Calu-3 cells to form a “tight monolayer”, the cultures growing on 3 m pores reached only significant TEER values after 5–7 days. We could not find literature data reporting TEER values using membranes with 3 m pores. The TEER values using 0.4 m pores considerably differ between different research groups. Mathias et al. (2002) reported a high value of >800 cm2 from day 8 until 16, using a collagen-coated air interface culture. The cultures of Foster et al. (2000) reached a TEER value of ∼600 cm2 after 10 days; Florea et al. (2001) reported a value of 440 ± 25 cm2 . It has been suggested that these differences may reflect the differences in the method of TEER measurement (Foster et al., 2000). However, culture conditions (density, medium, coating, . . .) also have a considerable impact. Moreover, the relatively high standard deviations in our experiments indicate that there is variability between monolayer TEER from culture to culture. It is, therefore, important to assess the TEER of the monolayers before each translocation experiment. Culture conditions have a considerable impact on the airway epithelial cells in vitro, hence it is crucial
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
to optimize these conditions for each cell type. In this respect, primary type II pneumocytes are more difficult to handle, because of the variability between different isolations. Growing primary type II pneumocytes on 0.4 m pores, a TEER value of 241 ± 90 cm2 was found on day 5 after isolation, which was significantly higher than the value on day 3. Two kinds of cell densities were tested, 0.3 × 106 and 0.6 × 106 cells/cm2 , with no important differences in TEER value. Coating with mono-fibrillar collagen did not improve the tightness of the monolayers. According to the high standard deviation, the TEER values were influenced by the isolation. Until now we have not succeeded to reach acceptable “tight” cell layers of type II pneumocytes on membranes with 3 m pores. Also, for the primary type II pneumocytes we could not find literature data on transepithelial electrical resistance in which membranes with 3 m pores were used. For the 0.4 m pore sizes, Sugahara et al. (1984) and Cott et al. (1986) had similar results although their isolation method and culture conditions were different. Cheek et al. (1989) and Kim et al. (1991) obtained higher TEER values of more than 2000 cm2 . Due to structural arrangement of the lungs the TEER value cannot be measured in vivo, however Dickinson et al. (1996) estimated the in vivo TEER value of the alveolar epithelium to be 2000 cm2 , which is higher than the TEER values in our experiments. Our results show that the Calu-3 cell line and the primary type II pneumocytes are suitable for use in an in vitro biphasic model with a 0.4 m pore size permeable membrane. Such a biphasic model can be used for translocation of chemicals, but not for particles, because the particles do not even cross the membrane. The Calu3 cell line has reached high TEER values on membranes with 3 m pores. Still, it is important to find a good condition to grow the primary type II cells on permeable membranes with 3 m pores. 4.2. Translocation studies To study translocation of nanoparticles, we used fluorescent amine-modified and carboxyl-modified polystyrene latex beads with a diameter of 46 nm. These non-toxic particles offer the possibility to evaluate surface chemistry modifications. For the translocation study, we used a non-fluorescent buffer, which has no auto-fluorescence at the wavelengths we used (Hamoir et al., 2003). Adding 1% BSA to the non-fluorescent buffer prevented aggregation of the particles. Due to the small size of the particles it is difficult to find a suitable method for detection. For example, it was not possible to detect low concentrations of the particles with spectrofluorime-
225
try. Therefore, we used fluorescence correlation spectroscopy for detection and quantification of the particles. The presence of fluorophore-labelled particles resulted in extremely high peaks of fluorescence, which cannot be analysed by means of autocorrelation. To analyse the fluorescence fluctuations the method of Van Craenenbroeck et al. (2001) was used. This method discriminates the fluorescence intensities originating from free fluorophores or from bright spikes. With this method, the detection limit was 0.04 g/ml. First, we examined the translocation of particles in inserts without a cell monolayer. Using 0.4 m pores, the translocation proved very small. Using 3 m pore size inserts 67.5% and 52.7% translocation of carboxylmodified and amine-modified particles, respectively, was observed, so these inserts are suitable to use for translocation studies. Using the 0.4 m pores there was no signal detected in the basolateral compartment with the Calu-3 and primary type II pneumocytes. This can be expected because the low passage with the control inserts. Using the 3 m pores there was a large difference between the different Calu-3 monolayers used. As shown if Fig. 5, TEER values did increase in most cultures during incubation with non-fluorescent buffer. In general, we found that low TEER values (lower than the “minimum TEER” as discussed before) were linked to a higher degree of translocation. High TEER values resulted in most cases in a lower degree of translocation. In these experiments, there was no difference in passage between carboxyl-modified and amine-modified particles. These data are preliminary and must be interpreted carefully. In conclusion, we noted a good correlation between the TEER value and the leakage of sodium fluorescein. The Calu-3 cell line and primary type II pneumocytes reached high TEER values on 0.4 m pore filters. Finding the right condition for cell culture is extremely important, even subtle changes can have an effect on the tightness of the monolayer. The translocation of particles through permeable membranes without a cell monolayer leads to the observation that 0.4 m pores are too small to use for translocation studies, 3 m pores are big enough. Until now, only the Calu-3 cell line showed a high resistance on membranes with a 3 m pore size, and therefore, can be used in an in vitro translocation study.
Acknowledgement This study was supported by a grant from FWOFlanders (G.0169.04).
226
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
References Agu, R.U., Jorissen, M., Willems, T., Augustijns, P., Kinget, R., Verbeke, N., 2001. In-vitro nasal drug delivery studies: comparison of derivatised, fibrillar and polymerised collagen matrix-based human nasal primary culture systems for nasal drug delivery studies. J. Pharm. Pharmacol. 53, 1447–1456. Cheek, J.M., Kim, K.J., Crandall, E.D., 1989. Tight monolayers of rat alveolar epithelial cells: bioelectric properties and active sodium transport. Am. J. Physiol. 256, C688–C693. Colvin, V., 2003. The potential environmental impact of engineered nanomaterials. Nat. Biotechnol. 21, 1166–1170. Cott, G.R., Sugahara, K., Mason, R.J., 1986. Stimulation of net active ion transport across alveolar type II cell monolayers. Am. J. Physiol. 250, C222–C227. Dickinson, P.A., Evans, J.P., Farr, S.J., Kellaway, I.W., Appelqvist, T.P., Hann, A.C., Richards, R.J., 1996. Putrescine uptake by alveolar epithelial cell monolayers exhibiting differing transepithelial electrical resistances. J. Pharm. Sci. 85, 1112–1116. Donaldson, K., Stone, V., Tran, C.L., Kreyling, W., Borm, P.J., 2004. Nanotoxicology. Occup. Environ. Med. 61, 727–728. Florea, B.I., van der Sandt, I.C.J., Schrier, S.M., Kooiman, K., Deryckere, K., de Boer, A.G., Junginger, H.E., Borchard, G., 2001. Evidence of P-glycoprotein mediated apical to basolateral transport of flunisolide in human broncho-tracheal epithelial cells (Calu-3). Br. J. Pharmacol. 134, 1555–1563. Foster, K.A., Avery, M.L., Yazdanian, M., Audus, K.L., 2000. Characterization of the Calu-3 cell line as a tool to screen pulmonary drug delivery. Int. J. Pharm. 208, 1–11. Foster, K.A., Oster, C.G., Mayer, M.M., Avery, M.L., Audus, K.L., 1998. Characterization of the A549 cell line as a type II pulmonary epithelial cell model for drug metabolism. Exp. Cell Res. 243, 359–366. Foster, K.A., Yazdanian, M., Audus, K.L., 2001. Microparticulate uptake mechanisms of in-vitro cell culture models of the respiratory epithelium. J. Pharm. Pharmacol. 53, 57–66. Hamoir, J., Nemmar, A., Halloy, D., Wirth, D., Vincke, G., Vanderplasschen, A., Nemery, B., Gustin, P., 2003. Effect of polystyrene particles on lung microvascular permeability in isolated perfused rabbit lungs: role of size ans surface properties. Toxicol. Appl. Pharmacol. 190, 278–285. Hoet, P.H., Lewis, C.P., Dinsdale, D., Demedts, M., Nemery, B., 1995. Putrescine uptake in hamster lung slices and primary cultures of type II pneumocytes. Am. J. Physiol. 269, L681–L689. Hoet, P.H., Nemmar, A., Nemery, B., 2004. Health impact of nanomaterials? Nat. Biotechnol. 22, 19. Kim, K.J., Cheek, J.M., Crandall, E.D., 1991. Contribution of active Na+ and Cl− fluxes to net ion transport by alveolar epithelium. Respir. Physiol. 85, 245–256. Luther, W., Nass, R., Campbell, R., Dellwo, U., Schuster, F., Tenegal, F., Kallio, M., Lintunen, P., Salata, O., Remˇskar, M., Zumer,
M., Hoet, P.H., Br¨uske-Hohlfeld, I., Lipscomb, S., Malanowski, N., Zweck, A., 2004. Technological Analysis, Industrial Application Of Nanomaterials—Chances and Risks. Future Technologies Division, D¨usseldorf. Mathias, N.R., Timoszyk, J., Stetsko, P.I., Megill, J.R., Smith, R.L., Wall, D.A., 2002. Permeability characteristics of Calu-3 human bronchial epithelial cells: in vitro–in vivo correlation to predict lung absorption in rats. J. Drug Target. 10, 31–40. Nemmar, A., Hoet, P.H., Vanquickenborne, B., Dinsdale, D., Thomeer, M., Hoylaerts, M.F., Vanbilloen, H., Mortelmans, L., Nemery, B., 2002a. Passage of inhaled particles into the blood circulation in humans. Circulation 105, 411–414. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Dinsdale, D., Smith, T., Xu, H., Vermylen, J., Nemery, B., 2002b. Ultrafine particles affect experimental thrombosis in an in vivo hamster model. Am. J. Respir. Crit. Care. Med. 166, 998–1004. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Nemery, B., 2004. Possible mechanisms of the cardiovascular effects of inhaled particles: systemic translocation and prothrombotic effects. Toxicol. Lett. 149, 243–253. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Vermylen, J., Nemery, B., 2003. Size effect of intratracheally instilled particles on pulmonary inflammation and vascular thrombosis. Toxicol. Appl. Pharmacol. 186, 38–45. Nemmar, A., Vanbilloen, H., Hoylaerts, M.F., Hoet, P.H., Verbruggen, A., Nemery, B., 2001. Passage of intratracheally instilled ultrafine particles from the lung into the systemic circulation in hamster. Am. J. Respir. Crit. Care Med. 164, 1665–1668. Oberd¨orster, G., 2001. Pulmonary effects of inhaled ultrafine particles. Int. Arch. Occup. Environ. Health 74, 1–8. Oberd¨orster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R., Kreyling, W., Cox, C., 2004. Translocation of inhaled ultrafine particles to the brain. Inhal. Toxicol. 16, 437–445. Oberd¨orster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R., Lunts, A., Kreyling, W., Cox, C., 2002. Extrapulmonary translocation of ultrafine carbon particles following whole-body inhalation exposure of rats. J. Toxicol. Environ. Health 65, 1531–1543. Sugahara, K., Freidenberg, G.R., Mason, R.J., 1984. Insulin binding and effects on glucose and transepithelial transport by alveolar type II cells. Am. J. Physiol. 247, C472–C477. Van Craenenbroeck, E., Vercammen, J., Matthys, G., Beirlant, J., Marot, C., Hoebeke, J., Strobbe, R., Engelborghs, Y., 2001. Heuristic statistical analysis of fluorescence fluctuation data with bright spikes: application to ligand binding to the human serotonin receptor expressed in Escherichia coli cells. Biol. Chem. 382, 355– 361. Winton, H.L., Wan, H., Cannell, M.B., Gruenert, D.C., Thompson, P.J., Garrod, D.R., Stewart, G.A., Robinson, C., 1998. Cell lines of pulmonary and non-pulmonary origin as tools to study the effects of house dust mite proteinases on the regulation of epithelial permeability. Clin. Exp. Allergy 28, 1273–1285.
In vitro study of the pulmonary translocation of nanoparticles A preliminary study J. Geys a , L. Coenegrachts a , J. Vercammen b , Y. Engelborghs b , A. Nemmar a , B. Nemery a , P.H.M. Hoet a,∗ a
Laboratory of Pneumology, Unit of Lung Toxicology, K.U. Leuven, Herestraat 49, Leuven 3000, Belgium b Laboratory of Biomolecular Dynamics, K.U. Leuven, Celestijnenlaan 200D, Leuven 3001, Belgium Received 20 May 2005; received in revised form 13 July 2005; accepted 13 July 2005 Available online 30 August 2005
Abstract Recent studies indicate that inhaled ultrafine particles can pass into the circulation. To study this translocation in an in vitro model three types of pulmonary epithelial cells were examined. The integrity of the cell monolayer was verified by measuring the transepithelial electrical resistance (TEER) and passage of sodium fluorescein. TEER was too low in A549 cells. In these preliminary experiments, TEER values of 1007 ± 300 and 348 ± 62 cm2 were reached for the Calu-3 cell line, using permeable membranes of 0.4 and 3 m pore size, respectively. Growing primary rat type II pneumocytes on 0.4 m pores, a TEER value of 241 ± 90 cm2 was reached on day 5; on 3 m pores, no acceptable high TEER value was obtained. Translocation studies were done using 46 nm fluorescent polystyrene particles. When incubating polystyrene particles on membranes without a cellular monolayer, significant translocation was only observed using 3 m pores: 67.5% and 52.7% for carboxyl- and amine-modified particles, respectively. Only the Calu-3 cell line was used in an initial experiment to investigate the translocation: on 0.4 m pores no translocation was observed, on 3 m pores ∼6% translocation was observed both for carboxyl- and amine-modified particles. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Rat type II pneumocytes; Calu-3 cell line; In vitro; TEER; Nanoparticles
1. Introduction Because of their physico-chemical properties, nanostructured materials (<0.1 m) can significantly differ from the corresponding coarse bulk material. The dosimetric aspects of the deposition and disposition of particles are different for inhaled ultrafine particles compared with larger particles (Oberd¨orster, 2001). Nanomaterials potentially have important industrial advantages. Concerns have been expressed that these properties also
∗
Corresponding author. E-mail address: [email protected] (P.H.M. Hoet).
might have adverse health and environmental impacts (Donaldson et al., 2004; Colvin, 2003; Hoet et al., 2004). The effect of nanostructured materials on biological systems has been studied mainly in view of the health risks posed by inhaling particles originating from air pollution: PM10 (particulate matter <10 m aerodynamic diameter), PM2.5 and PM0.1 (ultrafine particles ‘UFP’, diameter <0.1 m). Recent studies indicate that ultrafine particles can pass into the circulation (Nemmar et al., 2001). Although still somewhat controversial, it is accepted that some types of ultrafine particulates, such as elemental carbon, can be translocated to the blood circulation with subsequent uptake in the liver by 1 day after inhalation
0378-4274/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2005.07.005
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
exposure in rats (Oberd¨orster et al., 2002). Nemmar et al. (2002a, 2004) also showed translocation of ultrafine 99m Tc-carbon particles to the blood in healthy human volunteers. Recently, Oberd¨orster et al. (2004) demonstrated translocation of ultrafine particles to the brain via the olfactory nerve. The physico-chemical properties of the particulate surface play an important role in the biological effects in the lung and systemic circulation (Hamoir et al., 2003). Unmodified and negatively charged UFPs had no effect on thrombus formation whereas positively charged UFPs enhanced thrombus formation when administrated intravenously (Nemmar et al., 2002b) or instilled intratracheally (Nemmar et al., 2003). The Royal Society and Royal Academy of Engineering (RS/RAE 2004) recommend that industry should assess the risk of releasing products and materials containing nanoparticles or nanotubes throughout the lifecycle of the product and make this information available to the relevant regulatory authorities. The potential hazard of ultrafine particles and fibres is not predictable ‘a priori’ by their bulk physicochemical properties. Because in vivo experiments are expensive, slow and ethically questionable there is a strong demand for low-cost high-throughput in vitro assays without reducing the efficiency and reliability of the risk assessment (Luther et al., 2004). An in vitro cell culture model of the respiratory epithelium that closely mimics the human respiratory epithelium would be a valuable tool to screen ultrafine particles (or nanoparticles) for their capability to translocate through the pulmonary epithelium and to study this phenomenon in detail. Several cell lines are currently used as in vitro models of respiratory epithelium, e.g. A549 and Calu-3. The main purpose of this article is: (a) to assess the suitability of an in vitro model for studying the translocation of ultrafine particles across the respiratory epithelium and (b) to present preliminary data on the translocation of ultrafine model particles with different surface chemistry.
219
polystyrene carboxyl-modified fluorescent labelled particles (L5030, L3030) and sodium fluorescein were purchased from Sigma NV/SA (Bornem, Belgium). Deoxyribonucleic (DNase) I was obtained from Roche Molecular Biochemicals (Mannheim, Germany); Vitrogen collagen (3.0 mg/ml) from Nutacon (Leimuiden, The Netherlands) and Nembutal® (pentobarbital 60 mg/ml) from CEVA Sant´e Animal (Brussels, Belgium). Culture flasks were purchased from Iwaki Glass, International Medical (Brussels, Belgium); nylon filters (40 m) from Falcon, Becton Dickinson (Leuven, Belgium). Transwell Polycarbonate Membrane® and Transwell-Clear Polyester Membrane® inserts were obtained from Costar Europe (The Netherlands). The components of phosphate buffered saline (PBS− ) are 130 mM NaCl; 5.2 mM KCl; 10 mM glucose; 10.6 mM Hepes and 2.6 mM Na2 HPO4 ; pH 7.4; PBS+ contains in addition: 1.9 mM CaCl2 and 1.29 mM MgSO4 ; “non-fluorescent buffer” is composed of 118 mM NaCl; 4.75 mM KCl; 2.53 mM CaCl2 ·2H2 O; 2.44 mM MgSO4 ; 1.19 mM KH2 PO4 ; 25 mM NaHCO3 (Hamoir et al., 2003). 2.2. Cell lines
2. Materials and methods
The A549 (human alveolar epithelial cell line) and Calu-3 (human bronchial epithelial cell line) cell lines were obtained from ATCC and maintained as a monolayer culture in plastic tissue culture flasks. For the A549 cell line a Waymouth medium with 10% fetal calf serum, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM lglutamine and 1.25 g/ml fungizone was used. The Calu-3 cell line was maintained in MEM with 10% fetal calf serum, 1% non-essential amino acids, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM l-glutamine and 1.25 g/ml fungizone. The cells were incubated at 37 ◦ C, in a 100% humidified atmosphere containing 5% CO2 . When confluent, the cell lines were released enzymatically (0.1% trypsin–EDTA in PBS− ) and transferred to different culture plates. The medium was renewed every 2–3 days.
2.1. Materials
2.3. Isolation of primary type II pneumocytes
Waymouth medium, minimal essential medium (MEM), fetal calf serum, fungizone, l-glutamine (200 mM), penicillin–streptomycin (10,000 U/ml and 10,000 g/ml), non-essential amino acids and trypsin–EDTA were obtained from Invitrogen (Merelbeke, Belgium). Trypsin type I (T-8003), bovine serum albumin (BSA), Percoll, polystyrene amine-modified fluorescent labelled particles (L0780, L9529, L0155),
The method of Hoet et al. (1995) was used to isolate primary rat type II pneumocytes. Briefly, 150–200 g, male Wistar rats were anaesthetized with an intraperitoneal injection of Nembutal® (pentobarbital 1.5 ml/kg). The rats were exsanguinated and then tracheostomized. The lungs were perfused with 0.9% NaCl via the pulmonary artery. Finally, the heart was removed and the lungs together with the trachea were cut off. The lungs
220
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
were lavaged five times with 5–6 ml of 0.9% NaCl and then once with trypsin (250 mg in 100 ml PBS+ , pH 7.4). Then the lungs were filled every 5 min with trypsin and incubated at 37 ◦ C for 30 min. The trachea and bronchi were removed and the lungs were cut for 5 min until fragments smaller than 1 mm3 were obtained. Fetal calf serum (5 ml) and DNase I (1 mg) were added and the suspension was shaked for 5 min. After filtering through gauze and a 40 m nylon filter, the cell suspension was centrifuged for 30 min at 250 × g on a discontinuous Percoll gradient (ρ = 1.089 and 1.040 g/ml). The band on the interface was collected and pelleted by centrifugation for 10 min at 250 × g. The pellet of cells was resuspended in Waymouth medium with 10% fetal calf serum, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM l-glutamine and 1.25 g/ml fungizone and two times incubated for 40 min (37 ◦ C, 5% CO2 ) on a 60 mm petri dish. After centrifuging the cells, they were resuspended in the medium and counted using a B¨urker chamber. 2.4. Cell culture The cells were seeded onto Transwell Clear Polyester inserts of 0.33 cm2 or 1 cm2 , with pores of 0.4 m or 3 m diameter. Some experiments were done with Transwell Polycarbonate Membrane. The inserts were used uncoated, coated with mono-fibrillar collagen (Vitrogen100 diluted five times in PBS+ dried overnight) or coated with collagen gel (Vitrogen-100, 10× PBS− , pH 7.4, 60 min gelation at 37 ◦ C). Different seeding densities were used: 0.075, 0.15, 0.30, 0.45, 0.60, 0.75 and 0.90 × 106 cells/cm2 for the A549 cell line; 0.30, 0.60 and 1.20 × 106 cells/cm2 for the Calu-3 cell line and 0.30, 0.60 and 0.90 × 106 cells/cm2 for the primary type II cells. The cell monolayer was microscopically examined. 2.5. Measuring the integrity of the monolayer 2.5.1. Measurement of transepithelial electrical resistance Transepithelial electrical resistance was measured, after refreshing the culture medium, using MillicellERS (Millipore, Bedford, MA). Each sample was measured twice and the mean value was calculated. The mean resistance of a cell-free Transwell insert was subtracted from the resistance measured across each cell layer to yield the TEER value of the cell monolayer. The TEER of the Calu-3 cell line was measured daily from day 3 after seeding the cell culture until day 7; the type II pneumocytes were checked on days 3 and 5.
2.5.2. Sodium fluorescein transport Sodium fluorescein (1 mg/ml in non-fluorescent buffer) was added in the apical compartment, nonfluorescent buffer in the basolateral compartment. After placing the cultures for 60 min in the CO2 incubator (5% CO2 , 37 ◦ C) samples were taken from the basolateral compartment and the amount of sodium fluorescein was measured by spectrophotometry at 486 nm. The concentration in the samples was calculated using a standard curve with known concentrations of sodium fluorescein (10, 5, 2.5, 1.25, 0.625 and 0.3125 g/ml). The sodium fluorescein leakage was measured on the last day of the culture. A monolayer was considered to be a “tight monolayer” when the amount of sodium fluorescein in the basolateral compartment was less than 1% of the initial amount in the apical compartment. 2.6. Translocation studies In the basolateral compartment, the medium was removed and the compartment was filled with nonfluorescent buffer with 1% bovine serum albumin. The apical medium was replaced by medium without serum, containing 1% BSA and particles (25 g/ml). Two kinds of fluorophore-labelled polystyrene particles were used: amine-modified and carboxyl-modified polystyrene beads with a diameter of 0.046 m. Prior to use, the particle stock solution was sonicated for 15 min in a sonication-bath (Branson 1200, Branson Ultrasonics Corporation) and vortexed to reduce nonspecific aggregation. Translocation studies with Calu3 cells (0.60 × 106 cells/cm2 ) were performed on day 3 (0.4 m pores) and day 5 (3 m pores). Translocation studies with the primary type II pneumocytes (0.60 × 106 cells/cm2 ) were done on 0.4 m pores, on day 5. After 14–16 h of incubation, the medium in the apical and basolateral compartment was collected and the concentration of particles was measured. The concentration of the particles in the apical and basolateral compartment was assessed by fluorescence correlation spectroscopy (FCS) using a Confocor I (Zeiss Gmbh, Jena, Germany) confocal microscope. The 488 nm line of the Argon ion laser (power of 10 mW) was used for excitation and the emission of the fluoro-labelled particles was detected with an Avalanche Photodiode. The pinhole diameter was 60 m and the instrument was calibrated with Rhodamine 6G. The fluorescent molecules were measured in an open confocal volume with a typical size of 1.4 fl. Measurements were performed 10 times during 60 s (sample rate = 5 ms). To analyse the fluorescence fluctuations the method of Van Craenenbroeck et al. (2001) was used. This method dis-
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
221
Fig. 1. Transepithelial electrical resistance of Calu-3 cell monolayers. Calu-3 cells are seeded on Transwell Clear inserts with (A) 0.4 m pore size or (B) 3 m pore size; 0.3 × 106 cells/cm2 () or 0.6 × 106 cells/cm2 (). Data represent the mean TEER values ± S.D.; n = 4–3; three to two experiments (0.3 × 106 cells/cm2 ); n = 10; four experiments (0.6 × 106 cells/cm2 ). Student t-test between different seeding densities at the same day after seeding. ** p < 0.01 and * p < 0.05.
criminates the fluorescence intensities originating from free fluorophores or from bright spikes, using the statistical software S-Plus (Mathsoft Inc.). The intensities of the peaks of fluorescence were counted and the mean and standard deviation of the 10 measurements was derived. The concentration of the particles was calculated using a standard curve (5, 1, 0.2, 0.04 g/ml; four measurements each). 2.7. Statistical analyses Results are expressed as mean ± S.D. The data were analysed with paired or unpaired t-test and one-way ANOVA followed by Tukey’s multiple-comparison test (GraphPad Prism Package, GraphPad Software, San Diego, CA). p < 0.05 are considered significant. 3. Results 3.1. In vitro model A549 cells were seeded at different seeding densities on membranes with pore diameters of 0.4 m (0.075, 0.15, 0.30, 0.45, 0.60, 0.75 and 0.90 × 106 cells/cm2 ) and 3 m (0.15 and 0.45 × 106 cells/cm2 ), with or without collagen mono-fibrillar coating (n = 2 in two experiments). In all conditions, the TEER values did not exceed 25 cm2 and the permeability of sodium fluorescein was the same as that of the cell-free control inserts (data not shown). Seeding Calu-3 cells on 0.4 m pore size Transwell Clear inserts, mean TEER values of 404 ± 83 and 1007 ± 300 cm2 were reached on the third day for seeding densities of 0.30 × 106 and 0.60 × 106 cells/cm2 , respectively (Fig. 1A). After day 3, the TEER value declined. Calu-3 cells seeded on 3 m pore size membranes reached the highest
TEER value on day 5 (0.60 × 106 cells/cm2 ) or day 7 (0.30 × 106 cells/cm2 ) (Fig. 1B). The yield of primary type II pneumocytes was 4.3 ± 2.7 × 106 cells/rat. The viability, assessed with trypan blue exclusion, was more than 98%, and the purity, before plating just after isolation, was more than 85%. The cells were seeded on day 0 and the TEER was measured from day 3 on. The measured TEER values with different experimental conditions are presented in Table 1. Cells seeded on the 0.4 m pores formed a monolayer with the highest TEER value on day 5 after seeding. On the membrane with 3 m pores, primary rat type II pneumocytes did not develop a “tight monolayer”. When using mono-fibrillar collagen coating, the TEER values were very low. Cells seeded on collagen gel did not form a monolayer, the TEER values were as low as the control inserts without a cell layer (data not shown). Some experiments were also done using Transwell Polycarbonate Membrane® inserts. The TEER values in these experiments were lower than the TEER values in experiments using Transwell Clear Polyester Membrane® inserts (data not shown). 3.2. Integrity of the monolayer 3.2.1. Sodium fluorescein leakage The permeability of the monolayers was checked using sodium fluorescein. Plotting the TEER value and the permeability of sodium fluorescein against each other on a logarithmic scale, a linear correlation was found, both for the type II pneumocytes and the Calu3 cell line. This correlation is shown in Fig. 2 for the type II pneumocytes. According to Agu et al. (2001), we considered a monolayer with a leakage of less than 1% sodium fluorescein as a “tight monolayer”. Using the correlation between sodium fluorescein leakage and TEER a minimum TEER value for a “tight
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
222 Table 1 TEER values of type II pneumocytes
Day 3
Day 5
69 ± 23 cm2 (n = 11/5) isolations 24 ± 15 cm2 (n = 10/5)
224 ± 51 cm2 (n = 11/5) 89 ± 49 cm2 (n = 10/5)
0.4 m, 0.60 × 106 cells/cm2 No coating Coatinga
78 ± 46 cm2 (n = 25/11) 40 ± 21 cm2 (n = 12/11)
241 ± 90 cm2 (n = 25/11) 168 ± 56 cm2 (n = 12/11)
3 m, 0.60 × 106 cells/cm2 No coating Coatinga
7 ± 7 cm2 (n = 8/4) 10 ± 7 cm2 (n = 8/4)
31 ± 13 cm2 (n = 8/4) 25 ± 10 cm2 (n = 8/4)
3 m, 0.90 × 106 cells/cm2 No coating Coatinga
8 ± 3 cm2 (n = 10/4) 13 ± 4 cm2 (n = 8/4)
24 ± 9 cm2 (n = 10/4) 30 ± 6 cm2 (n = 8/4)
0.30 × 106
0.4 m, No coating Coatinga
cells/cm2
n = #/# isolations. a Mono-fibrillar collagen coating.
Fig. 2. Relation permeability sodium fluorescein and TEER value. Type II pneumocytes are seeded on 0.4 m pore size Transwell Clear inserts without mono-fibrillar collagen coating; n = 37. (A) Relation between permeability sodium fluorescein and TEER values. (B) Logarithmic transformation of both parameters and linear regression (r2 = 0.89; p < 0.0001). The line points to a leakage of 1% sodium fluorescein, with which the corresponding TEER value can be calculated.
monolayer” was calculated. The minimal TEER values for Calu-3 cells seeded on 0.4 and 3 m pores were 575 and 420 cm2 , respectively. The minimal TEER value of primary type II pneumocytes seeded on 0.4 m pores without mono-fibrillar collagen coating was 251 cm2 and with mono-fibrillar collagen coating 304 cm2 . 3.3. Translocation studies 3.3.1. Influence of changing the cell culture medium on TEER Changing the culture medium of the primary type II cells to non-fluorescent buffer had no effect on the TEER value (n = 5, see Fig. 3C). Incubating the culture with carboxyl-modified or amine-modified particles lowered
the TEER value of primary type II cells, but not significantly (n = 5, see Fig. 3C). After 14–16 h of incubation of the Calu-3 monolayer with particles, the TEER values were significantly higher under all conditions compared to before incubation (n = 6, see Fig. 3A and B). 3.3.2. Translocation through inserts without cells The distribution of the particles, after 14–16 h incubation, between the apical and the basolateral compartment of an insert cell culture system was first examined without the presence of a cell monolayer. The results are shown in Fig. 4. The translocation of carboxyl-modified particles was 13.5% through 0.4 m pores (n = 7) and 67.5% through 3 m pores (n = 3). The translocation of amine-modified particles was 4.2% through 0.4 m pores (n = 7) and 52.7% through 3 m pores (n = 3).
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
223
Fig. 3. Influence of changing the cell culture medium on TEER. Calu-3 cells are seeded on a (A) 0.4 m pores (n = 6) and (B) 3 m pores (n = 6); primary type II pneumocytes are seeded on 0.4 m pores (C) (n = 5). Cell monolayers are incubated with 25 g/ml carboxyl-modified particles () or amine-modified particles (), or non-fluorescent buffer (䊉). Data represent mean ± S.D. *** p < 0.001, ** p < 0.01 and * p < 0.05.
3.3.3. Translocation through a cell monolayer Because the TEER values of the A549 culture were too low, no translocation studies were done with this cell line. The translocation studies were performed using the Calu-3 cell line (0.4 and 3 m pore size) and primary type II cells (0.4 m pore size). Incubating Calu-3 cell monolayers with particles (25 g/ml) on the 0.4 m pore size membrane, we could not detect particles in the basolateral compartment. Fig. 5 shows the individual data of six different cultures of Calu-3 cell monolayers growing on 3 m pore size membranes incubated overnight with 46 nm particles. A limited fraction of the particles, both carboxyl- or amine-modified particles (25 g/ml), incubated at the apical side of the cultures translocated to the basolateral compartment. We could not detect
any translocation in 2 (of 5) and 3 (of 6) experiments with carboxyl- or amine-modified particles, respectively. After incubation, 1.05 (±0.29) g particles and 1.16 (±0.19) g particles were in the apical compartment of the carboxyl-modified particles and amine-modified particles, respectively. In the basolateral compartment 0.06 (±0.05) g particles were found for both particle types. This represents, on average, a translocation of ∼6% of the particles from the apical to the basolateral compartment. After incubating the primary type II pneumocytes monolayer with particles on the 0.4 m pore size membrane, we could not detect any translocation to the basolateral compartment (data not shown). Because of too low TEER values no translocation studies were done with 3 m pore size.
Fig. 4. Translocation of particles through inserts without a cell monolayer. Carboxyl-modified or amine-modified particles were introduced at the apical compartment on inserts without a cell monolayer. The concentration of particles in both compartments after 14–16 h incubation is measured. (A) carboxyl-modified particles; 0.4 m pore size, n = 7; (B) carboxyl-modified particles; 3 m, n = 3; (C) amine-modified particles; 0.4 m, n = 7; (D) amine-modified particles, 3 m, n = 3. Data represent mean ± S.D.
224
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
Fig. 5. Translocation of particles through a Calu-3 monolayer. Calu-3 cells are seeded on a 3 m pore membrane and incubated with 25 g/ml carboxyl-modified particles (A) or amine-modified particles (B). Per monolayer culture three values are given: the first two (connected with full line) present the TEER value before and after the incubation with particles. These two values have to be read from the left y-axis. The third value (most right), presents the total (g) mass of particles translocated into the basolateral compartment (n = 6).
4. Discussion Recent studies indicate that inhaled ultrafine particles can pass into the circulation (Nemmar et al., 2001, 2002a; Oberd¨orster et al., 2002). Moreover, the physicochemical properties of the particulate surface plays an important role in the biological effects in the lung and systemic circulation (Hamoir et al., 2003). At this moment the mechanisms and kinetics of particle translocation are not yet known. Therefore, the aim of this study was to build an in vitro model, including the setup of a suitable cell culture system, to study this translocation phenomenon. This can be a tool in screening the toxicity and bioavailability of different materials before marketing. Moreover, such a model can also be a valuable tool for pharmacological studies. 4.1. Integrity of cell monolayer In developing an in vitro model it is important that it reflects the in vivo situation as closely as possible. At this moment such a system has not yet been described. We evaluated three types of cell monolayers to mimic the in vivo respiratory epithelial barrier: the commercially available and widely used A549 and Calu-3 cell lines, as well as primary rat type II pneumocytes, which are more cumbersome to obtain. In general, a monolayer is said to be a “tight monolayer” if the leakage of sodium fluorescein is less than 1% (Agu et al., 2001). However, the disadvantage of sodium fluorescein is the inability to use the culture afterwards. We noted a good correlation between the TEER value and the leakage of sodium fluorescein. Thanks to this correlation it is possible to monitor the tightness of the cell layer, using TEER, before (and after) use in an experiment.
In our experiments, we were unable to obtain high TEER values with the A549 cell line. This is consistent with the reported inability of the A549 cell line to form tight junctions (Winton et al., 1998). Also Foster et al. (1998, 2001) reported low TEER values using the A549 cell line. This indicates that the A549 cell line is not suitable for studying translocation. The present study confirms previous reports indicating that the Calu-3 cell line is capable of forming tight monolayers in culture, thus generating high TEER values. Working with the Calu-3 cell line, TEER values of 1007 ± 300 and 348 ± 62 cm2 were obtained using permeable membranes with 0.4 and 3 m pore size, respectively. A large pore size reduces significantly the capacity of the Calu-3 cells to form a “tight monolayer”, the cultures growing on 3 m pores reached only significant TEER values after 5–7 days. We could not find literature data reporting TEER values using membranes with 3 m pores. The TEER values using 0.4 m pores considerably differ between different research groups. Mathias et al. (2002) reported a high value of >800 cm2 from day 8 until 16, using a collagen-coated air interface culture. The cultures of Foster et al. (2000) reached a TEER value of ∼600 cm2 after 10 days; Florea et al. (2001) reported a value of 440 ± 25 cm2 . It has been suggested that these differences may reflect the differences in the method of TEER measurement (Foster et al., 2000). However, culture conditions (density, medium, coating, . . .) also have a considerable impact. Moreover, the relatively high standard deviations in our experiments indicate that there is variability between monolayer TEER from culture to culture. It is, therefore, important to assess the TEER of the monolayers before each translocation experiment. Culture conditions have a considerable impact on the airway epithelial cells in vitro, hence it is crucial
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
to optimize these conditions for each cell type. In this respect, primary type II pneumocytes are more difficult to handle, because of the variability between different isolations. Growing primary type II pneumocytes on 0.4 m pores, a TEER value of 241 ± 90 cm2 was found on day 5 after isolation, which was significantly higher than the value on day 3. Two kinds of cell densities were tested, 0.3 × 106 and 0.6 × 106 cells/cm2 , with no important differences in TEER value. Coating with mono-fibrillar collagen did not improve the tightness of the monolayers. According to the high standard deviation, the TEER values were influenced by the isolation. Until now we have not succeeded to reach acceptable “tight” cell layers of type II pneumocytes on membranes with 3 m pores. Also, for the primary type II pneumocytes we could not find literature data on transepithelial electrical resistance in which membranes with 3 m pores were used. For the 0.4 m pore sizes, Sugahara et al. (1984) and Cott et al. (1986) had similar results although their isolation method and culture conditions were different. Cheek et al. (1989) and Kim et al. (1991) obtained higher TEER values of more than 2000 cm2 . Due to structural arrangement of the lungs the TEER value cannot be measured in vivo, however Dickinson et al. (1996) estimated the in vivo TEER value of the alveolar epithelium to be 2000 cm2 , which is higher than the TEER values in our experiments. Our results show that the Calu-3 cell line and the primary type II pneumocytes are suitable for use in an in vitro biphasic model with a 0.4 m pore size permeable membrane. Such a biphasic model can be used for translocation of chemicals, but not for particles, because the particles do not even cross the membrane. The Calu3 cell line has reached high TEER values on membranes with 3 m pores. Still, it is important to find a good condition to grow the primary type II cells on permeable membranes with 3 m pores. 4.2. Translocation studies To study translocation of nanoparticles, we used fluorescent amine-modified and carboxyl-modified polystyrene latex beads with a diameter of 46 nm. These non-toxic particles offer the possibility to evaluate surface chemistry modifications. For the translocation study, we used a non-fluorescent buffer, which has no auto-fluorescence at the wavelengths we used (Hamoir et al., 2003). Adding 1% BSA to the non-fluorescent buffer prevented aggregation of the particles. Due to the small size of the particles it is difficult to find a suitable method for detection. For example, it was not possible to detect low concentrations of the particles with spectrofluorime-
225
try. Therefore, we used fluorescence correlation spectroscopy for detection and quantification of the particles. The presence of fluorophore-labelled particles resulted in extremely high peaks of fluorescence, which cannot be analysed by means of autocorrelation. To analyse the fluorescence fluctuations the method of Van Craenenbroeck et al. (2001) was used. This method discriminates the fluorescence intensities originating from free fluorophores or from bright spikes. With this method, the detection limit was 0.04 g/ml. First, we examined the translocation of particles in inserts without a cell monolayer. Using 0.4 m pores, the translocation proved very small. Using 3 m pore size inserts 67.5% and 52.7% translocation of carboxylmodified and amine-modified particles, respectively, was observed, so these inserts are suitable to use for translocation studies. Using the 0.4 m pores there was no signal detected in the basolateral compartment with the Calu-3 and primary type II pneumocytes. This can be expected because the low passage with the control inserts. Using the 3 m pores there was a large difference between the different Calu-3 monolayers used. As shown if Fig. 5, TEER values did increase in most cultures during incubation with non-fluorescent buffer. In general, we found that low TEER values (lower than the “minimum TEER” as discussed before) were linked to a higher degree of translocation. High TEER values resulted in most cases in a lower degree of translocation. In these experiments, there was no difference in passage between carboxyl-modified and amine-modified particles. These data are preliminary and must be interpreted carefully. In conclusion, we noted a good correlation between the TEER value and the leakage of sodium fluorescein. The Calu-3 cell line and primary type II pneumocytes reached high TEER values on 0.4 m pore filters. Finding the right condition for cell culture is extremely important, even subtle changes can have an effect on the tightness of the monolayer. The translocation of particles through permeable membranes without a cell monolayer leads to the observation that 0.4 m pores are too small to use for translocation studies, 3 m pores are big enough. Until now, only the Calu-3 cell line showed a high resistance on membranes with a 3 m pore size, and therefore, can be used in an in vitro translocation study.
Acknowledgement This study was supported by a grant from FWOFlanders (G.0169.04).
226
J. Geys et al. / Toxicology Letters 160 (2006) 218–226
References Agu, R.U., Jorissen, M., Willems, T., Augustijns, P., Kinget, R., Verbeke, N., 2001. In-vitro nasal drug delivery studies: comparison of derivatised, fibrillar and polymerised collagen matrix-based human nasal primary culture systems for nasal drug delivery studies. J. Pharm. Pharmacol. 53, 1447–1456. Cheek, J.M., Kim, K.J., Crandall, E.D., 1989. Tight monolayers of rat alveolar epithelial cells: bioelectric properties and active sodium transport. Am. J. Physiol. 256, C688–C693. Colvin, V., 2003. The potential environmental impact of engineered nanomaterials. Nat. Biotechnol. 21, 1166–1170. Cott, G.R., Sugahara, K., Mason, R.J., 1986. Stimulation of net active ion transport across alveolar type II cell monolayers. Am. J. Physiol. 250, C222–C227. Dickinson, P.A., Evans, J.P., Farr, S.J., Kellaway, I.W., Appelqvist, T.P., Hann, A.C., Richards, R.J., 1996. Putrescine uptake by alveolar epithelial cell monolayers exhibiting differing transepithelial electrical resistances. J. Pharm. Sci. 85, 1112–1116. Donaldson, K., Stone, V., Tran, C.L., Kreyling, W., Borm, P.J., 2004. Nanotoxicology. Occup. Environ. Med. 61, 727–728. Florea, B.I., van der Sandt, I.C.J., Schrier, S.M., Kooiman, K., Deryckere, K., de Boer, A.G., Junginger, H.E., Borchard, G., 2001. Evidence of P-glycoprotein mediated apical to basolateral transport of flunisolide in human broncho-tracheal epithelial cells (Calu-3). Br. J. Pharmacol. 134, 1555–1563. Foster, K.A., Avery, M.L., Yazdanian, M., Audus, K.L., 2000. Characterization of the Calu-3 cell line as a tool to screen pulmonary drug delivery. Int. J. Pharm. 208, 1–11. Foster, K.A., Oster, C.G., Mayer, M.M., Avery, M.L., Audus, K.L., 1998. Characterization of the A549 cell line as a type II pulmonary epithelial cell model for drug metabolism. Exp. Cell Res. 243, 359–366. Foster, K.A., Yazdanian, M., Audus, K.L., 2001. Microparticulate uptake mechanisms of in-vitro cell culture models of the respiratory epithelium. J. Pharm. Pharmacol. 53, 57–66. Hamoir, J., Nemmar, A., Halloy, D., Wirth, D., Vincke, G., Vanderplasschen, A., Nemery, B., Gustin, P., 2003. Effect of polystyrene particles on lung microvascular permeability in isolated perfused rabbit lungs: role of size ans surface properties. Toxicol. Appl. Pharmacol. 190, 278–285. Hoet, P.H., Lewis, C.P., Dinsdale, D., Demedts, M., Nemery, B., 1995. Putrescine uptake in hamster lung slices and primary cultures of type II pneumocytes. Am. J. Physiol. 269, L681–L689. Hoet, P.H., Nemmar, A., Nemery, B., 2004. Health impact of nanomaterials? Nat. Biotechnol. 22, 19. Kim, K.J., Cheek, J.M., Crandall, E.D., 1991. Contribution of active Na+ and Cl− fluxes to net ion transport by alveolar epithelium. Respir. Physiol. 85, 245–256. Luther, W., Nass, R., Campbell, R., Dellwo, U., Schuster, F., Tenegal, F., Kallio, M., Lintunen, P., Salata, O., Remˇskar, M., Zumer,
M., Hoet, P.H., Br¨uske-Hohlfeld, I., Lipscomb, S., Malanowski, N., Zweck, A., 2004. Technological Analysis, Industrial Application Of Nanomaterials—Chances and Risks. Future Technologies Division, D¨usseldorf. Mathias, N.R., Timoszyk, J., Stetsko, P.I., Megill, J.R., Smith, R.L., Wall, D.A., 2002. Permeability characteristics of Calu-3 human bronchial epithelial cells: in vitro–in vivo correlation to predict lung absorption in rats. J. Drug Target. 10, 31–40. Nemmar, A., Hoet, P.H., Vanquickenborne, B., Dinsdale, D., Thomeer, M., Hoylaerts, M.F., Vanbilloen, H., Mortelmans, L., Nemery, B., 2002a. Passage of inhaled particles into the blood circulation in humans. Circulation 105, 411–414. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Dinsdale, D., Smith, T., Xu, H., Vermylen, J., Nemery, B., 2002b. Ultrafine particles affect experimental thrombosis in an in vivo hamster model. Am. J. Respir. Crit. Care. Med. 166, 998–1004. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Nemery, B., 2004. Possible mechanisms of the cardiovascular effects of inhaled particles: systemic translocation and prothrombotic effects. Toxicol. Lett. 149, 243–253. Nemmar, A., Hoylaerts, M.F., Hoet, P.H., Vermylen, J., Nemery, B., 2003. Size effect of intratracheally instilled particles on pulmonary inflammation and vascular thrombosis. Toxicol. Appl. Pharmacol. 186, 38–45. Nemmar, A., Vanbilloen, H., Hoylaerts, M.F., Hoet, P.H., Verbruggen, A., Nemery, B., 2001. Passage of intratracheally instilled ultrafine particles from the lung into the systemic circulation in hamster. Am. J. Respir. Crit. Care Med. 164, 1665–1668. Oberd¨orster, G., 2001. Pulmonary effects of inhaled ultrafine particles. Int. Arch. Occup. Environ. Health 74, 1–8. Oberd¨orster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R., Kreyling, W., Cox, C., 2004. Translocation of inhaled ultrafine particles to the brain. Inhal. Toxicol. 16, 437–445. Oberd¨orster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R., Lunts, A., Kreyling, W., Cox, C., 2002. Extrapulmonary translocation of ultrafine carbon particles following whole-body inhalation exposure of rats. J. Toxicol. Environ. Health 65, 1531–1543. Sugahara, K., Freidenberg, G.R., Mason, R.J., 1984. Insulin binding and effects on glucose and transepithelial transport by alveolar type II cells. Am. J. Physiol. 247, C472–C477. Van Craenenbroeck, E., Vercammen, J., Matthys, G., Beirlant, J., Marot, C., Hoebeke, J., Strobbe, R., Engelborghs, Y., 2001. Heuristic statistical analysis of fluorescence fluctuation data with bright spikes: application to ligand binding to the human serotonin receptor expressed in Escherichia coli cells. Biol. Chem. 382, 355– 361. Winton, H.L., Wan, H., Cannell, M.B., Gruenert, D.C., Thompson, P.J., Garrod, D.R., Stewart, G.A., Robinson, C., 1998. Cell lines of pulmonary and non-pulmonary origin as tools to study the effects of house dust mite proteinases on the regulation of epithelial permeability. Clin. Exp. Allergy 28, 1273–1285.