Air–Liquid Interface Exposure of Lung Epithelial Cells to Low Doses of Nanoparticles to Assess Pulmonary Adverse Effects

Reliable and predictive in vitro assays for hazard assessments of manufactured nanomaterials (MNMs) are still limited. Specifically, exposure systems which more realistically recapitulate the physiological conditions in the lung are needed to predict pulmonary toxicity. To this end, air-liquid interface (ALI) systems have been developed in recent years which might be better suited than conventional submerged exposure assays. However, there is still a need for rigorous side-by-side comparisons of the results obtained with the two different exposure methods considering numerous parameters, such as different MNMs, cell culture models and read outs. In this study, human A549 lung epithelial cells and differentiated THP-1 macrophages were exposed under submerged conditions to two abundant types of MNMs i.e., ceria and titania nanoparticles (NPs). Membrane integrity, metabolic activity as well as pro-inflammatory responses were recorded. For comparison, A549 monocultures were also exposed at the ALI to the same MNMs. In the case of titania NPs, genotoxicity was also investigated. In general, cells were more sensitive at the ALI compared to under classical submerged conditions. Whereas ceria NPs triggered only moderate effects, titania NPs clearly initiated cytotoxicity, pro-inflammatory gene expression and genotoxicity. Interestingly, low doses of NPs deposited at the ALI were sufficient to drive adverse outcomes, as also documented in rodent experiments. Therefore, further development of ALI systems seems promising to refine, reduce or even replace acute pulmonary toxicity studies in animals.


Cell-free DCF test
The test uses the oxidation of the non-fluorescent 2´,7´-dichlorodihydrofluorescein (DCFH2) to the fluorescent 2´,7´-dichlorofluorescein (DCF) as an indicator for the presence of reactive oxygen species [1][2] and has been performed as described previously [3]. Briefly, DCFH2-diacetate was deacetylated with NaOH by mixing 0.1 mL of 5 mM DCFH2-DA (Invitrogen, Karlsruhe, Germany) in ethanol with 2.4 mL of 0.01 N NaOH and incubating at room temperature (24°C) for 30 min. For neutralization, 10 mL PBS was added and kept on ice in the dark until use. Just prior to use, horseradish peroxidase (HRP, Sigma, Taufkirchen, Germany) was added as a catalyst (2.2 U/mL). The DCFH2 concentration in the working solution was 40 µM.
Suspensions of test particles were prepared in PBS (10 mg/mL), sonicated for 10 min and further diluted, and H2O2 standard preparations (0.04 to 10 µM) were also prepared. The test solutions were mixed 1:1 (v/v) with the prepared DCFH2 solution and incubated at 37°C for 15 min in the dark. Then, solutions were centrifuged (20,000 × g for 15 min) to remove the particles and the fluorescence of the supernatant was monitored at 485 nm excitation and 530 nm emission using a fluorescence microplate reader (BIO-TEK FL600 from MWG-Biotech AG, Ebersberg, Germany). Results were expressed as fold changes relative to the particle free sample.

Culture of THP-1 cells and differentiation to macrophages
The human myeloid leukemia cell line THP-1 was obtained from DSMZ (German Collection of Microorganisms and Cell Cultures, Braunschweig, ACC 16). Cells were cultured in RPMI 1640 medium containing 10 % FBS, 2 mM L-glutamine, 100 µg/mL penicillin, and 100 U/mL streptomycin. The non-adherent THP-1 monocytes were differentiated into macrophage-like cells by treatment with 30 ng/mL TPA (12-O-tetradecanoylphorbol-13-acetate, Sigma, Taufkirchen) for 4 days and incubation in TPA-free medium for 3 days [4]. TPA-differentiated THP-1 cells (dTHP-1) resemble some biological and morphological characteristics of human alveolar macrophages, such as eicosanoid and cytokine production [5]. The dTHP-1 cells become adherent and no longer divide. The differentiation was confirmed by detection of CD14 expression using flow cytometry.

MCP-1 release
Secreted monocyte chemotactic protein-1 (MCP-1) was analyzed in the cell culture medium using the MCP-1 ELISA kit from eBioscience (Frankfurt, Germany) according to the manufacturer's instructions.
For measurement of absorption and data analysis, a microplate reader and the software package SoftMaxPro (Molecular Devices, Ismaning, Germany) were used.

Gene expression and genotoxicity studies under submerged conditions
TiO2 NPs were pre-wetted with ethanol and suspended in 0.5 mg/mL sterile BSA at the concentration of 5 mg TiO2/mL according to the NanoGenoTox protocol [6]. Immediately before cell exposure, they were sonicated for 15 min at 70% of amplitude, at 4°C, using an indirect cup-type sonicator (cup-horn), operated via a Vibracell 75041 sonicator (Fisher Bioblock, Rungis, France). They were then diluted in cell culture medium (DMEM without FBS) and applied to cells.

Analysis of p53 binding protein 1 (53BP1) foci
DNA double strand breaks (DSB) or replication fork blockade were determined by counting p53 binding protein 1 (53BP1) foci in cell nuclei. 53BP1 is a non-enzymatic protein which is recruited shortly after primary DSB detection. This protein is homogeneously distributed in the nuclei of unperturbed cells and it is recruited within 1-2 min to DSB sites [7]. Like gamma-H2AX it can therefore serve as a marker for DSB. The method has been performed as described previously [8]. As a positive control, cells were exposed to 25 µM etoposide (Sigma-Aldrich) for 24 h. Briefly, cells were fixed for 20 min in 3% paraformaldehyde (Sigma-Aldrich), stained using anti-53BP1 antibody (Novus Biologicals, Littleton, CO, USA, 1/500 vol./vol.) and slides were mounted with Fluoroshield (Sigma-Aldrich) containing DAPI (1 slide per condition). 53BP1 foci were visualized on an Axio ImageA1 microscope coupled to an Axiocam MRm camera (Carl Zeiss). At least 15 images per condition were captured; on each image both, total number of 53BP1 foci and total number of nuclei, were determined. Apoptotic cells and dividing cells were rejected.  [10], e BET determined by the producer, f BET determined by different laboratories [9], g determined by different laboratories [9], h determined by TEM by different laboratories [9].  Table S1) whereas for ceria NPs surface area was calculated based on the diameter as listed in Table 1 Figure S1. Aerosol generation from NP suspensions. The continuously stirred suspension of NPs is dispersed via a two-phase nozzle into the aerosol reactor. Inside the reactor the aerosol is dried and is led to the automated exposure station. Figure S2. The unmodified and redox-modified CeO2 NPs have no effect on viability of THP-1 macrophages but slightly increase IL-8 release. The cells were seeded as mono-culture and exposed to CeO2-NPs in medium without serum for 24h. LDH release (a), AlamarBlue assay (b), IL-8 (c) and MCP-1 release (d) were performed as described in Figure S2. The positive control was exposed to 0.1 µg/mL LPS. The results are means ± s.e.m. of two independent experiments performed in duplicate. *p < 0.05, **p < 0.01 and ***p < 0.001 indicate significant differences of treated cells compared to control cells exposed to medium only.  and alkali-labile sites as shown in Figure 6. However, under submerged conditions strand breaks are only provoked at much higher doses yet no enhanced 53BP1 foci formation could be observed. (a) At the ALI, the cells were exposed to clean air or to TiO2-NPs at the indicated doses but for different time periods (0.17 µg/cm²: 30 min aerosol, 0.17* µg/cm²: 30 min aerosol + 3 h 30 min air, 1.14 µg/cm²: 4 h aerosol) and 53BP1 foci were monitored. Cells were also exposed under submerged conditions (b, c) and the alkaline comet assay was performed (positive control: 50 µM H2O2) (b), or 53BP1 foci were analyzed (positive control: 50 µM etoposide) (c). Data are means of three independent experiments ± SD. Statistics *p < 0.05, exposed vs. control (clean air (ALI) or unexposed cells (submerged)).