Silica nanoparticles are less toxic to human lung cells when deposited at the air–liquid interface compared to conventional submerged exposure

Summary Background: Investigations on adverse biological effects of nanoparticles (NPs) in the lung by in vitro studies are usually performed under submerged conditions where NPs are suspended in cell culture media. However, the behaviour of nanoparticles such as agglomeration and sedimentation in such complex suspensions is difficult to control and hence the deposited cellular dose often remains unknown. Moreover, the cellular responses to NPs under submerged culture conditions might differ from those observed at physiological settings at the air–liquid interface. Results: In order to avoid problems because of an altered behaviour of the nanoparticles in cell culture medium and to mimic a more realistic situation relevant for inhalation, human A549 lung epithelial cells were exposed to aerosols at the air–liquid interphase (ALI) by using the ALI deposition apparatus (ALIDA). The application of an electrostatic field allowed for particle deposition efficiencies that were higher by a factor of more than 20 compared to the unmodified VITROCELL deposition system. We studied two different amorphous silica nanoparticles (particles produced by flame synthesis and particles produced in suspension by the Stöber method). Aerosols with well-defined particle sizes and concentrations were generated by using a commercial electrospray generator or an atomizer. Only the electrospray method allowed for the generation of an aerosol containing monodisperse NPs. However, the deposited mass and surface dose of the particles was too low to induce cellular responses. Therefore, we generated the aerosol with an atomizer which supplied agglomerates and thus allowed a particle deposition with a three orders of magnitude higher mass and of surface doses on lung cells that induced significant biological effects. The deposited dose was estimated and independently validated by measurements using either transmission electron microscopy or, in case of labelled NPs, by fluorescence analyses. Surprisingly, cells exposed at the ALI were less sensitive to silica NPs as evidenced by reduced cytotoxicity and inflammatory responses. Conclusion: Amorphous silica NPs induced qualitatively similar cellular responses under submerged conditions and at the ALI. However, submerged exposure to NPs triggers stronger effects at much lower cellular doses. Hence, more studies are warranted to decipher whether cells at the ALI are in general less vulnerable to NPs or specific NPs show different activities dependent on the exposure method.


Dynamic light scattering
Particles were suspended in deionized water at 10 mg/ml, shortly vortexed and probe sonified with 30 strokes, 50% cycle duty, output control: 8 (Branson Sonifier 250, Schwäbisch Gmünd, Germany). This suspension was further diluted in deionized H 2 O to 1 mg/ml. For DLS analysis, SiO 2 -NPs were further diluted to 50 µg/ml in deionized H 2 O or DMEM without serum. The samples were either analysed directly (0h) or after incubation at 37°C and 5% CO 2 for 24 h immediately after vortexing using the Zetasizer Nano ZS (Malvern Instruments Ldt., Herrenberg, Germany) at 25°C.

Determination of the deposited mass dose for Aerosil200 particles after ALI exposure
Determination of the mass dose for Aerosil200 particles is complex since they are not only composed of a broad size distribution of agglomerates but these agglomerates are also composed of monomers with a broad size distribution. Furthermore, the agglomerates are not as compact as those composed of the SiO 2 -50nm particles. If we analyse the TEM micrographs loaded with Aerosol200 particles in the same way as for the compact SiO 2 -50nm agglomerates the corresponding deposited mean mass dose is about 78 µg cm -2 .
However, in this case this value is an upper limit for the actual mass dose since the Aerosil200 agglomerates show a more complex cluster structure compared to SiO 2 -50nm agglomerates that cannot be described by a packing of hard spheres. In fact the effective densities of Aerosil200 agglomerates decrease with increasing agglomerate size. We used an aerosol particle mass analyser (APM) in combination with a differential mobility analyser (DMA) to determine the effective density of the Aerosol200 agglomerates as a function of their mobility equivalent size. As shown in Figure S1 this dependence can be well described by a bi-exponential function. Figure S1: Effective densities measured for Aerosol200 dispersed with an atomizer by combination of a differential mobility analyser (DMA) and an aerosol particle mass analyser (APM). The observation is well represented by a bi-exponential function.
Since 90% of the detected particles on TEM micrographs were smaller than 800 nm in projected area equivalent diameter and the decrease in effective density becomes small for large particles we estimate a lower limit for the real mass dose to about 26 µg cm -2 by using an average density of 0.4 g cm -3 . A more accurate estimation however requires a better understanding of the correlation between the projection equivalent diameter, mobility equivalent diameter and effective density and has to be part of future work. Nevertheless, the real mass dose has to be between the limits determined above and can be estimated to (52 ± 26) µg cm -2 .  Table S1). Based on DLS measurements of the SiO 2 -50nm particle suspension, we estimate that 90% of the mass is present in the fraction of the small particles and 10% in the fraction of the large agglomerates.

Deposition kinetics of the mass doses for
During ALI exposure the deposited dose increases linearly until 5 h (Aerosil200) and 7 h (SiO2-50 nm) and remains constant during the post-incubation period until 24 h. (Figure S2- The dose values were taken from Table 1 and the way how the dose was determined is described in the text.
Under submerged conditions the cellular doses of Aerosil200 and SiO2-50 nm increase continuously over 24 h as shown in Figure S2-A and in a magnification of the lower part of the diagram in Figure S2-B. For both particles the calculated cellular dose reaches 7 µg cm -2 after 24 h.